Loading...
HomeMy WebLinkAboutAPA3342--, ~ ._, -, _jj -, ..... ~ ~· ..4 -" _jj · SUJVll\~;H ·:.;CQJ.,OGY AND m<::HAVIOR Ol' TilE GRAYLING OF lVIcl/Ii!..IIDS CitEEIC AI,ASKA A 'l~liT.SIS l Presented to the l~culty of the University of Alaska in Partial Fulfi~lment of the Requirements for the Degree of l\'IAS~r:E!R OJ~· 8C.IEI\fCl:~ By Gian L. "Jascotto, ..,...,. C." _t). '--t. ARDIS Alask~:Re~urces · Library & Inr~rY.atton Services Anchor~~ Alaska · ,., College, Alaska J,':a~,r 197 0 / / \ANCHORAGE, ALASKA t(t . 63? .'...-! ....-\ l :;) ~--~ I/ 2: ~;:? v .;:."" 0 ,.,, ~J / ... / :~/ ,'Y.vL-/ '-. Est 1997, • ./. UNIV. Of'AL.AS~lBRAR'tl -, ...., " N ,c.o iN ~0') _jo N ....,0 ·0 0 _il.!) 1.!) r--- 9('1) M _. ;,i; __;, _j --' SU~~R ECOLOGY AND BEHAVIOR OF THE GRAYLING OF McMANUS CREEK ALASKA A{'PROVED: • ·--.....----'---.-----..;.....-. ·~--· -- ·-'-:-·--- . !? ,. ' .·J r!----i:..d~~·~-----· ~ .. ?:.~ ~ . .e /~...:,_ __ _ 7".,.. . _.,, . Cha~· r .an . ~ c~,-:~~~.·~d .-/: ~~ .. ~--'~~L/ --/}" =-~~~~----~ APPROVED:~'O.. ·kkP_ Dean of the College of Biological Sciences and Renev-;able Resources DNI'E: J26:~1JMI-_6J ~~···~ Vice President for Research and Advanced Study ARLIS Alaska Resources Library & Information Services Anchorage, AlMka 1 ' -, --", _;; ~ __$ ~ --., --' ~ d _. -., ..-1 _;; J ABSTRACT -,. A study ofthe summer ecology and behavior of the Arctic Grayling was undertaken on the summer population of McManus Creek, Alaska. An.effort to determine the distribution and the natterns . 4 of the fish's movements in the stream was.made. The grayling spent the summer months in pools where they established feeding territories. Within each feeding territory a feeding range, where all feeding activities took place, was found. In each pool a hierarchial ordering based on a dominant subordinate relationship existed. This hierarchy -v;as established and maintained by a series of displays. The grayling of McManus Creek were found to feed solely on the surface and at mid-depth._ The food items consisted both of" flying insects and aquatic insects, the latter making up the largest portion of their diet. It appeared that the fish relied primarily upon benthic drift for nutrition. Being.visual feeders, the fish were unable to utilize the large numbers of organisms known to drift during periods of high and muddy water. iii --, " ...., -~ ..., .., -" :..ii _i! -, =<I _;; -" ACKl.JO WLEDGElViENrr:s Research on l\1d1anus Creek, Alaska·1 was fj_rw.nced by grants from the Institute of Water Resources Research of the University of Alaska to D1'. James E. IvT.orrow, ProfE:E\f>or of Zoology, Department of Biological Sciences • I would like to express my appreciation to Dr. Korrow, Chairman of my Committee, for hi.s help ana guidance through- out the study and for his critical.!.. reading of the manuBcript. Mrs. Judith S. Weeden, Lecturer in Zoology, Department of Biological Sciences, and Dr. George C. West, ProfeE!<:~or o·P Zoophysiology, Institute of Arctic Biology, critically read the manuscript. Thanks also are due Dr. l·. Gerard Swa~t:JJ, Associate Professor of Zoology, Department of Biological Scier~ces, who identified a sample of Nematomorphs. l':t:r·. :Yynn Boddi.e, graduate student in fisheries biology, ar.,d Kr. Paul J. Frey of the Alaska Water IJabora"tory contril~ut,:'}d to some valuable discussions. --, --.. d -, _. .... _. -' TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . PAGE 1 INTRODUC'I'ION STUDY AREA •• . . . . . . . . . . . . . . . . • • ·-4 HABITAT, SEASOl\TAL MOVEJ:.1ENTS AND ABUNDANCE . . . . . • . 9 MATERIALS AND METHODS • . . . . . . . . . . . • • • 9.·. RESULTS • . . . . . . . . . . . . . . . . ~ . .12 Early summer .• ~ . . . . . • .12 ~ Mid summer. . . ~ . . . . . . . . . . . . .15 Late summer Fall •• Tagging results •• Characteristics of pools. DISCUSSION ••••• . . . . . . . . . . . . . . .17 . . 17 .19 . . .21 ")? • ·• e .. - DISTRIBUTION, TERRITORIALITY, AND SOCIAL HIEHARCHY • • • 29 MATERIALS AND METHODS . . . . •-. ~ . . . """9 . /(. RESULTS • . . . . . . . . . . . . . . . Horizontal and vertical distributiQn .. Feeding range and feeding center. . . . • . 32 . .32 . .44 Territories and hierarch:i.es • • DISCUSSION. . . . . . . . . . . . . . . . . FOOD HABITS. • • . . . . . . . . . . . ~ ~ . . MATERIALS AND r>~IETHODS • • • • • • " • • 4 • RESULTS • . . . . . . . . . . . . . . . . . . . . Utilization of food in the habitat .. . . . . Aerial organisms. . . . . .. . . . .. . . ,. .11-g .60 .70 .70 • 75 • 75 • 811- DISCUSSION. ~ •••.•• . . . . . . . . . .. . . • .$7 v '·• PAGE 96 ~~- FEEDING BEHAVIOR . . . . . . . . . . . . . . . . . . . . . ' METHOD OF STUDY . . . .. . • . . . . . . . . . . . . 96 -., RESULTS 98 . . . . . . . . . . . . . . . . . . . . . . . Group A . . . . . . . . . . . . . • . . . . . . . 98 ' ( ·. Group B . . . . . . . . • . . . . . . . . . . . . 101 Group c . . . . . . . . . . . . . . . . . . . . . 105 Feeding intensities . . . . . . . . . . . .. . . . .106 DISCUSSION . . . . • • . • . . . . . . . . . . . . 109 SUMMARY . . • . . . . . .. . . . . . . . . . . . . . . . 117 LITERATURE CITED . . . . . . . . . . . . . . . . . . . . 120 APPENDIX I . . . . . . . . . • . . . • . . . . . . . . . 124 ., "'"' APPENDIX II . . . . . . . . . . . . . . . . . . . . . . . 130 :" APPENDIX III . . . . • . . . . . . . . . . . . . . . . . 131 ....., -, ~ vi LIST OF TABLES TABLE PAGE ~. I. CHARACTERISTICS OF SIX OBSERVED POOLS • • • • • • 31 ( II. MOVEMENTS OF GRAYLING OVER TEN MINUTE PERIODS •• 47 -, III. RESULTS FROM ANALYSIS OF DRIFT SAMPLES . . . • . 89 IV. STOMACH·· CONTENTS ON RAINY DAYS • • • • • • . • • •.11"? ""' ~ r::- -· ~ -~ ! ~ ... """ ._J vii '" --, d --_, --., -· --' =" -~ _j LIST OF FIGURES FIGURE PAGE 1. Map of McManus Creek showing the h-3 pools studied and the three main sections of stream • 5 2~ Map of McManus Creek showing the 24 seining 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. sections and the six observation pools . . . . . 6 Distribution of fish in early summer . . . . . 13 . " Distribution of fish.in mid summer . . . . . 1" • -0 Di~stribution of fish in late summer • . . . . . 18 Results of the recaptures of 18 fish • 20 Pool AA . . . . . . . . . . 33 Pool BB . . . . . • . . • . . 34 Pool BB 1 • • . . • • • Pool CC • . . . Pool cc 1 • • . . . . . . . . . . . . • • 3 5 . 36 . • • . 37 Vertical distribution of grayling in a pool 38 Distribution of fish in narrow pools • . • . • 41 14. Movements of fish in the pool below the dam •• 42 15. Tracings of the movements of two fish over 16. 17. 18. 19. ten minute periods . . . . . . . . . . . . Steps in establishing dominance in a set Steps in territorial defense 1:;.6 . 51 • • 56 Movements of fish in territorial defense .•. 58 Hypothetical hierarchial ordering of fish . . • 65 viii ~ ~ _. -, -~ -' ~ -, --, d ._i _j -" ~ FIGURE 20. LIST OF FIGURES Hypothetical distribution within a pool 21. Absolute abundance of organisms in the stream . . . . . . . . . . . . . . . . . . . 2.2. Relative abundance of organisms in the stream . . .. . . . . . . . . . . . . . . . .. l 23. Volumetric and numerical distribution of food organisms . . . . . . . . . 24. Numerical abundance of aquatic organisms in the stomach . . . . . . . . . 25. Volumetric abundance of aquatic organisms in the stomach . . . . . •· . . . . . . . 26. Frequency of occurrence of a.quatic organisms in the stomach . . . . . . . . . . . . . . . 27. Frequency of occurrence of aerialorganisms in the stomach PAGE 68 71 72 76 78 79 83 85 28. Relative importance of food organisms . • . . 90 29. Triangles obtained by plotting frequency, numerical values, and volumetric values . 30. 31 . Feeding movements by Group A Feeding movements by Group B . . 91 . 100 .• 104 ix ---,. --. ~ ~ ~' -~ r ;-· L r~ l~ ! L_. r~ l~ [ [ [ - l INTRODUCTION The Arctic Grayling Thymallus arcticus (Pallas), is one of the most important game fishes of interior Alaska. Little is known about the life history, ecology and behavior ( . . of this .species. A survey of the summer movements, distri- bution, feeding habits and behavior of the grayling made in ·an interior Alaska, non-glacial stream exposed new facts which could form a base for future studies. The Arctic Grayling is distributed in the Arctic watersheds from the Ob River in Russia east to the Churchill River on Hudson Bay, Athabaska and Churchill drainages in Alberta, anJ. established artificially in a few of the North Saskatchewan ri~er.systems (Carl, Clemens, and Lindsey, 1959). Some grayling survived the last glaciation south of the ice sheet in Michigan where they became extinct in the 1930's, and in the upper Missouri drainage of P.•Iontana. Cutting of the forest, and the subsequent scouring of the bottom by the logs decimated the spawning areas, and the stocking of rainbow trout apparently contributed to the extinction of the I'-'Iichigan grayling (Vincent; .1962). f'hYmallus arcticiis inay reach a length of 20 to 24 inches and weigh from 2 .to 5 pounds. It has a very dark back and . purplish-gray sides varying in darkness.. The 1 --, ---, """1 4 ~ <~ 0' <--"' __., ... "-" ~ ~ ' w 2 belly is whitish with gray blotches. A yellmv stripe runs along the ventral side from the pel vic' ,fins to the pectoral fins,in adults. The large dorsal fin is long and high, fts base longer than the depth of the body. It is dark gray to black and blue in color, punctuated with pink and azure ( spots and deep blue cross":'"rows, often bordered with red and yellow. The pelvic fins, abdominal irt,position, are crossed by three pink stripes. Tlie head is brownish in color, with large eyes, and a small terminal mouth. Brown (1938b), Leonard (1938), Miller (1946), and Schallock (1965), ptiblished notes on the food habits of the fish. They shoY.Jed that the grayling's diet is made up almost entirely of small insects and aquatic larvae. They, suggest that the diet is limited by their small mouth, weak teeth, and cold water habitat. The Arctic Grayling's habitat has not been extensively studied. In Alaska, the fish is described as inhabiting most of the fast clear streams of the Yukon basin (EvermB-nn and Goldsborough, 1906). In the winter grayling are also reported from the glacier-fed streams which during this season run clear. The grayling are also common in lak,es, < where some of the largest specimens have been reported. In lakes they favor the mouths of streams and the outlets of the lakes. Spa~;-tning has been observed in :Montana (Brovm 1938, Nelson 1952, and Tryon 19~t-7) where it is reported to -~ --.,. __.; WI __;; --.] · take place between }:larch 15 7 and June 1 a.t temperatures belmv 10° C. The Alaska spawning dates· given by '!tJarner (1958) and Reed (1964) are from May 20 to June 25. Wojcik (195.5) reports finding ripe individuals bet'J!Jeen May 13 and June 16. The spawning act is described in ( : . detail by Brown (l938a) and Reed (1964). Tryon (19L .. 7) a~d \liard (1951) describe them as polygamous non--nest builders. l The fish produce from 2000 to.4000 eggs, amber in color owing to the presence of oil drops which render them semi-buoyant; their adhesive nature enables them tb become coated ltd th stirred-up sand so that they vfill sink to the bottom (Brown, l938a). Warner (1958) did not find the eggs adhesive, and reported considerable numbers being washed dmvnstream. Incubation is complete after t-vvo weeks and varies 3 with the temperature of the water. Fry begin feeding on plankton at about the fourth day (Brovm and Buck, 1939). The European species has been studied much more thoroughly, especially in terms of managemeht, and substan- tial ethological and ecological data have been gathered by Degteva ( 1965 )'; Egorov ( 1956), Fabricius and Gustafson (1955), Kafanova (1965), Peterson (1968), Sommani (1953), Starmach (1956), and Svetovidov (1936). , , " ..., -, ~ .dl --"' ..... DESCRIPTION OF STUDY AREA The ·study \•ms conducted along 18 .miles oi' McManus Creek from its origin near Twelvemile Summit, altitude "3150 feet, to a diversion dam at mile 68, Steese Highway, altitude 1450 feet. The area is located between 146°00' 0 0 0 to 146 25' West Longitude and 65 17' to 65 25' North Latitude .(Figs._l and 2) in interior Alaska. McManus Creek is a youthful' rapid runoff stream, having a narrow V-shaped valley and a steep gradient. The · gradient drops 400 feet per mile at the headwaters and 50 feet per mile at the junction with Faith Creek, resulting in a grade of 7.7cfo and 0.9cfo respectively (Schallock, 1965) • Thirteen small tributaries enter the stream in a trellis pattern, the major ones being Montana Creek, Idaho Creek, Smith Creek, and Faith Creek. _Of the four, Faith Creek is the largest and nearly doubles the volume of McManus (Fig • l) • Half a mile below the junction of Faith Creek arid McManus Creek (Fig. 1), the flo"tr is interrupted by a dam built to divert water into a man-made ditch (Davidson Ditch) ·;for hydroelectric use. This dam delineated the lmver limit of the study area. Due to the presence of boulders, the coarseness of the -bottom, and the steepness of the gradient, the wateri are constantly being churned, resulting in excellent, aeration. 4 ~ ' -., --" ..,; .J -'1 .., ~- Figure 1. Map showing McManus Creek, the 43 pools studied, and the three sections of stream. Reproduced from U.S.G.S. Alaska Map, Circle (B-5) • II. --:z.--""'-- L. ;;;-0 .... -., -, _] __.j --" Figure 2. Map showing McManus Creek, the 24 seining sections~ and the six observation pools. Reproduced .from U.S. G. S. Alaska Map·, Circle ( B-5) • ~ . ----=z.__,__ '-- --, --, -' ~ -" • .J ~ -" 7 The water is exceptionally clear. Human pollution fs nil since there are no mining operations or inhabitants in the drainage except at mile 82 of the Steese High~t;ay where a seasonal highway camp is located. The character of the stream remains relat:i,vely constant from two miles from the stream's origin to the junction with Montana Creek. ·. Cascading riffles.· are predominant, l -intermixed with small pockets of slower v·:ater lying either by. the undercut banks or below boulders in the middle of the stream. Most of the "pools" are 12 to 24 inches deep and less than 30 feet long; often there are vvater chutes where the water deepens, resulting in stretches of slow water near the bottom. Deeper·pools are present at -the junctions of the various creeks. Below the mouth of Montana Creek (Fig. 1), the gradient decreases noticeably and the creek takes on a braided appearance. Larger stretches ofquiet water are present, the pools are larger and deeper with small sweepers commonly trailing in the '·Tater (Table I, Appendix I). Typically, the side of. the pool a\"!ay from the undercut bank consists o:f a rubble or gravel bar which is submerged during the spring. In the last five miles there are long stretches of qu:i_et, deep water. The surface is comnwnly shadowed by trees. The banks are high and rubble bars are rare. Deep holes a.re prese'nt on the outside of bends, with many --, ~1 --.., ...., ' ~;; _ _J ... _.; -' -" 8 uprooted trees lodged on the banks. Some of these hoJ..es reach a depth of over five feet, and the quiet waters extend for hundreds of feet. The water looks much darker, ·apparently from the organic· matter deposited on the bottom by the relatively weak current. y _j ' -' -' _..; ' ..... -' -" THE HABITAT, SEASONAL MOVEMENTS, AND ABUNDANCE OF GRAYLING MATERIALS AND METHODS In order to determine the type of habitat utilized by the grayling in McManus Creek, 43 pools known to contain grailing_were deicribed in detail. The size of each pool was estimated by the investigator after·having measured by tape five pools, which showed that the estimated size showed an error of less than 15~. · The depth, v'lfhenever possible, was measured by walking into the stream and reading the maximum depth on hip boots which had been marked off at six inch .intervals. Where the pools were too deep, the depth was estimated. The bottom composition of each pool was determined by a visual estimation. All rocks larger than 10 inches in diameter were considered boulders; all rocks between 1 and 10 inches as rubble; and any rocks between approximately 3/16 and 1 inch as gravel •. Everything smaller was considered as sand. Daily aj,r temperatures for the·area were obtained by .means of a PTC Il/fodel 615 dry stylus thermogra.ph, ·.located at the junction of Faith.and McManus Creeks. The instrument was wrapped in a plastic bag, placed inside a can nailed to ·a spruce tree five feet above the ground, and away from direct sunlight. 9 ~ L . ....J I L I I L- L ~ __. --" __. -" 10 The \Yater temperatures were taken by means of .a centigrade laboratory thermometer in the riffles above each pool. ~he thermometer was completely immersed and lodged between rocks for five minutes. At three stations marked I, II, and III on Figure 2, ( . 250 ml water samples were collected in airtight glass bottles, placed in a closed box, and kept. below 56°F. Later the contents were mixed in•a Waring Blender and turbidity readings taken on a Hellige 1urbidimeter. The pH was measured at the the three stations using Brom-Thymol Blue and a Bach colorimeter model DR-1599 B. Stream velocities were obtained with an irrilllersible Kahl Scientific Instrument Co. water flow meter. The meas- urements were taken from randomly selected pools in each of the 24 stations at mid-depth, in feet per second except where. stated otherwise. In order to determine the population of the stream and the .presence of seasonal movements, tagging operations were conducted throughout the summer of 1966 and 1967. The stream was divided into 24 stations (Fig. 2). The fish were captured with 1/8 and l/4 inch mesh 20 foot drag seines, and by hook and line. Each individual was anesthetized, fork length was measured, the weight determined withChatillon spring scales,. and for later studies scales were removed- from the right.side of the fish, just below the dorsal fin. --" ·--' .J ~ _;j 11 A monel :peduncle ta~ was affixed and the fish released. The fish were released in the sarD.e section where they v1ere captured, but not necessarily j_n the same pooi. At the pools marked AA, BB, CC, and DD (Fig. 1), 21 fish were tagged with color coded spaghetti tags for quic~ identification~ Using the four.color rings--yellow, red, blue, and green--and alternating their order, 21 different combinations were obta:i.ned. ~~~-,, K~ --, -, " '! __. ~ _. --., d _. _. _:. 12 . RESUL'rS EarlY summer -' No data were obtained during the spring months. The ice began to break up in the creek betv-1een May 20 and May 25, 1967; but the investigator did not reach the area until June 4. From June 12 to June 20, 1967, and from June 16 to June 2·4, 1968, grayling vlere see:q· attempting to surmount the diversion dam. Fishermen ha.d reported seeing fish jump prior to these dates. For descriptions of the jumping see Schallock (1965). ·Between 2145 and 2.300 on June 1.3, 1967, ten fish were counted trying to negotiate the dam; of these, only tvvo were successful. At this time there was a drop of 4.5 to 5.0 feet between the lip of the dam and the stream •. Between June 10 and June 20, 1967, the creek was seine.d from station 24 to station 6 (Fig. 1) to determine the fish population present. The water vvas high, swift, and carrying a considerable silt load. The banks consis.ted mostly of packed snow and ice· up to 10 feet thick, and in portions the ice nearly blocked the stream. During the ten-day seining operation, <:mly one grayling was capture~, this being a 120 mm-parr taken on June 12 (Fig . .3). The fish was found under brush protruding from ·the bank of a backwater pool. The pool was 27 feet long, L I " L l, 0 w ~10 .... 0. < u :!:: ..,. -... ... 0 5 ~ w r•l A ~ ~ 3 2 I I I I l ,J : j l, j 'J J 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1·5 16 17 18 19 20 21 22 23 2A STREAM STATIONS l. J Figure 3. Distribution of fish in early surn.mer, as determined by seining .from station 24 to station 6. f-l w -, . .., -" ...., ~ --" --, - _:j ~ 14 6 feet wide, and 3 feet deep, located approximately 100 yards below station 19 ._ Ice was present along the edges, and the . . . 0 bottom was sand-covered. The water temperature_was 4 C, and the velocity within the pool was zero feet per second. The curr,ent in the main channel was _measured at 3.5 fps at center; and-1.7 fps at one third of the distance from the bank. In the-sam~ c;:>peration two parr approximately 150 mm l long were-seen-between stations 22 and 14. In both cases the parr wq.s scurrying_under the ice in rapid current. A third fish approximately 250 mm long was seen moving upstream in the main channel in· a current of 6 fps. In the very shallow riffles, between rocks at the lm·rer end of rubble bars, large numbers_of grayling alevins under 40 mm long were found. ·They appeared to be basking in micro- pools about three inches in depth. In these pools the alevins moved together in school-like fashion. They ~ested close to the bottom, but if frightened they exhibited avoidance reactions by either moving out into faster water or hiding under the rocks. The water in these micro-pools reached temperatures up to 10° C during the day but dropped to 4° and 5° C at night. --, -, ..... , . ...ii ~ ·-· ~ =' _, --" -' '-' i5 :Mid summer The creek was seined from station· 24 to 0, between June 28 and July 15. By the latter date. the water level at Faith and McManus had dropped 16 inches and the turbidity had dropped to 4 parts per million of suspended matter. · The velocity at the point where the first parr had been captured was, at this time, down to 1.7 fps. On June 26, nine grayling were collected between stations 3 and 5, and. all were found in the faster stretches •. Two were seen attempting to surmount riffles and were· immediately caught. Three vvere captured near the banks, where the water was two to three ~eet ·deep and verJJ-swift (3.5 fps). The rerri.aining four.were taken from a deep sandy pool, six feet long and six feet wide and more than four feet deep, both upstream and downstream from Yihich the water was extremely fast and turbulent. Six of the nine fish had ragged tails and dorsal fins. Observations of the grayling's behavior were made at two pools. These were both twenty feet long, four to five feet deep and on the north bru1ks. Dead spruce trees were jutting into the current. On June 26 the pools held 20 to 25 parr, all under 180 mm in length. Occasionally large fi-sh were seen moving upstream into the pool and disappear- ing under the. sweeper. They came singly and in groups of two or three.. Tn that same after-11.oon, three individuals were seen leaving the pool, heading upstream. L L L ~ .J . J lJ l. . ; l ... ,J L ,,J j j I I ! 8 20 r- ~ I ::I .... c.. .:_r i I n ~10~ II)U ,)l ~ I j! ")';· I I 5 . I z I I I I l 01 2 3 4 5.6 7 8 9 101112 i31415 1617181920 212223 24 S iREAM STATIONS l ' Figure 4. Distribution of fish in mid-summer, as determined by seining from station 24 to station 0. '-' I C"· J ~ ' .., [ [ E Li I '-' L r - L 17 The seining operations from June 28 to July 15 indicated ·that the fish were distributed evenly through the length of the stream (Fig. 4). The seining began at station 24 and worked downstream. On the 28th, the maximum water temperature was 5° C and by the end of the seining opera- ( . . . 0 tions. the temperature re,ached a high of 13 C. Ice 1-vas present along stre.t~ches of the. banks until ,July 10. ' Late. -~.!!1~ Between August 25 and September 5, 1966, McManus "\'ras seined by graduate students Lynn Boddie and Marshall Danby • . Using the same sta~ions but a smaller mesh seine, the results indicated that the bulk of the. fish were occupying the upper half of the stream (Fig.· 5). Although no seining ws.s conducted in 1968, a distribution similar to Figure 6 was evident from my observations. Fall The data available for periods after Se~tember 5 are scanty; in 1968~ observations were made at three pools, one in station 22, one in station 18, and one in station 1, on September 1 and 14. In the pools of stations 18 and.22 the fish 1-vere still present and in larger concentrations than at any previous time. On August 20, grayling parr were attempt- ing repeatedly to_jump over the spillways of the culverts. "\'rhere Md<Ianus' tribi.1taries are crossed by the Steese High"·ray They were unsuccessful, since a drop of six feet was present. ' ' ' -. -"' ., 0 ... "' ~ ::;) ..... tl. = <( v X <.1'1 .... ' ..... 0 IX ... co ~ ~ :::::l z t...,.l I' k.. _. '--' :......~ ~-~ I 60 50·- 40 30 20 10 .. J _ __r- _L _____ ~------· 01234567 8 9 10 Hl21314 1516 17 18 19 20 21 22 23 24 Figure 5. STREAM STAHON S Distribution of fish in late surr~mer,. as determined by seining from station l to station 24. 18 -, -., ' .... ~ --' ~ ·-" __. ~~­ ~,;: ;:c 19 (j1 • .J..agg1ngrasu1ts In the su:mmer of 1966,.354 fish were tagged·a.nd 144 fin-clipped in Mc:M:anus Creek. At the time of tagging, the crew had estimated that at least 75% of the fish with fork lengths greater than 180 m.m ·had been captured due to the ( . low water and high visibility (Lynn Boddie~ personal communication). Such conditions would have increased the efficiency· of the seine by allovv:fng the workers to see the fish. Betwe.en September 28 and September 30, 1966, 18 fish which had been tagged on Au.§>'U.st 28, 1966 were recaptured by a fisherman (Fig. 6). Over the one month between tagging and recapture, the :fish exhibited a random upstream-dovmstre8.Jll movement. During the summer of 1967,-in two seining operations, 105 fishweretagged and 59 fin-clipped. During the same operation, 10 fish tagged in 1966, and 17 others vii th scars froiiilost tags were recaptured. Of the fish tagged in 1966 2. 8% were recaptured still bearing tags, and 7. 5~lfo bearing either tags or tag scars. Four tagged fish were recaptured below the dam. During 1968, . only five tagged fish were recaptured from the previous years. One of these had been tagged in 1966, the others in 1967. Two of. them were taken below the da:tn~ None were f6uncl with tag scars. -, ....., ...., -" c ~ ~ u " _. __. -' SECTION IN VIHICI-I FISH WEi!E RECAPTURED 7 7 7 7 7 7' 7 1 7 7 ~-~---. -.--. r-· 5 4 l 31111 m I ~ <( ~2 ... ~lillJlUIIIIIIIII nliJ ; -l - 1 ~ z . . z 2 1- w ~~ .... ><C 0"" ~!':31-· ..... z ~ 0 041- 5t- I I __l____!_ ~ 2 3 _4 5 6 7 8 9 10 SECTiON i.N WHICH FISH WEf:E TAGGE'D Figure 6. Results of the recapture of 18 fish,. tagged August 28-, 1966, and recovered September 28-30, 1966. 20 J J J J J J J J J J J J J J J J J J J 21 Of the fish with colored tags, four from pool AA vanished in the first week of July. Three of these were recovered in the third week of August bet·ween pools BB and CC. The fourth one disappeared and is presumed to have been captured by fishermen. The fate of some of ·the remaining 17 will be.discussed later. .Characteristics of pools in which ,the ~li~ were fopnd ' A' sample of 43 pools was studied· intensively to describe the type. of habitat utilized by the fish.· Of these, 18 pools were selected from the upper portion, 18 from the middle and 17 from the lower section of streai1l (Fig. 1). The lengths of the pools varied from six feet to 150 feet, the . average being 35 feet. The data obtained were·· tabulated in Table I, Appendix I, and the mean value3 per section for pool lengths, depths, surface area and estimate of plant cover are shown in Table II, Appendix I. The C!:ata on plant cover were converted to freq_uency of occurrence per section, and then to freq_uency of occurrence for the entire. study etrea. The results of the evaluati.on of the bottom composition, water velocities, .and surface are tabulated in Tables III and IV, Appendix I. J J J J J J J J J J J J J J J J J ] J 22 DISCUSSION The seining data indicate that, even allowing for loss of effectiveness of the seine due to the high and murky water, few fish were present in McManus Creek up to mid June.; All those seen or captured were small and probably survived the.wini;er in one of the deeper pools. The pr~sence of fry in the s~ream from Ju~e. 12 to June ' . 20,. 1967, and the absence of mature individuals at this time, appear to indicate that·the grayling which are repo·rted to spawn in late May and early June (Nelson 1954, RE?ed i964), had spawned in the stream at least two weeks earlier, and then had migrated back downstream. Such a movement would appear to agree with Reed (1964), Warner (1956), and.Ward (1951) who indicate that spawning grayling from lakes move up into the creeks in May or early Jm1.e, and then move back downstream within one or two weeks. In ]l[cfdanus Creek the grayling apparently move upstream to spawn as soon as the ice goes out. Once spent they move back downstream. A few weeks later, another group ready ~o spawn moves up and this is the population which inhabits the stream during the summer. This second.group would·be the fish in spavming ·condition which were captured below the diversion dam as late as June 20. Reed (1964), in J J J J J ] J J J J J J J ~J ] j "] J J 23 describing graylir~ streams of the Tanana drainage, mentions that some grayling remained in the stream after spawning; this does not seem to be the case with the grayling responsible for the mid-June fry in McManus Creek, but would apply to the later migrants. Another possibility would be that some grayling may spawn in. the fall, .with the fertilized eggs remaining in ·the gravel over the \''linter. Unfor~unately no ripe individuals have been seen by the investigator or other workers from the Alaska Department of Fish and Game in the fall. Fishermen from other drainages, notably the Salcha River, maintain that they have caught ripe individuals in late August. Schallock (1965) indicates that the young of the year vanish from the stream by August. This was not the .case i.n 1966, 1967, orl968. Both young of the year and one year old fish remained common throughout the summer. ;However, they did change their habits considerably, since in June and July they were commonly found in shallow pools formed by rubble bars. By A11ooust they were found in the deeper po.ols, where they stayed close to the brush and were camouflaged with the bottom. . Their. presence was shown during the tagging operations in 1966, ·when 95 fish under 180 mm were fin-clipped between stations 17 and 24 and countless ones under 95 mm we.re released,. all after August 25. Similar results were· obtained in 1967. l J J ] J J J J J J J J ] J ] J ] ~·· ;·;~,:. "{~ .-.-.··., ~j,. j~f~. 24 It app~"ars that most of the summer population of the stream comes from below the dam. The evidence for this can be seen in Figures 4 and 5; which show that in the early part of June the stream is nearly devoid of fish,. but is well populated by July le Further evidence to support the above . idea'is given by data from the recapture of grayling marked the pr~vious year. Even if only 5o% of .the taggable fish ·had been tagged.in 1966, the com:Qiete absence of fin-cli-pped recaptures in 1967, and· the low recapture, 7. 5%, of previo1.~sly tagged fish·strongly.implies an almost complete turnqver in population. ·The presence of previously tagged fish below the diversion dam; 18.8% of the total recaptures, where they were captured and seen attempt~ng to jump, leads to specu- lation that emigration in fall orwinterdoes take place. It should be noted that recaptured fish showed that the tag had severely damaged the fish's peduncular region. A marked swelling and.open wound was present on all recaptures, even though the tag had been inserted the previous year .. It appeared that the swimming was hampered, and such a wound could have severely affected the ability of the fish to negotiate.· the dam. The sharp decline in number of fish captured in 1967 (165), opposed to 498 in: 1966, may ·be a reflection of the damage caused by the tags either by keeping fish from surmount:i.P_g the dam, or by increasir-€ the mortality of the ·im1ividuals. 25 An upstream spring or early swr.rner migration is evj_dcnt in McMa·cms Creek. This mi.::;ratj_(m conforr:1s to the data available from other localities, Brorm (1938a) in 11ontana; Degteva (1965) in Russia; Leach (1923) and Vincent (1962) in r:Iichi.:_--;an; and Warner (1956) in Alaska. However the migration in this area differs from the others in that here the fish do not returr1 in~mediately downstrearn to a lake after spavvnj_nJ.·, e.ncl that tho T~cr:anus grayli:ag do not n;ove en masse, but either singly or in groups of two or three. This may be due to the presence of the diversion clo_r:t \'o,:hich disrupts the movement of the fish as a .;roup, or due to the high latitude, which results in virtualJy contin~ous daylight during mid and late J'une. Other authors report observinG the migra--'cion of the fish only after CW1dOVI:l. The pools occupied in early sUID .. n~er are those found in the deepest and slovvest portion of the strec:.rn, whj_ch also contains the largest pools. ·rhese characteristics are found in the loVler section of the stream (Tables II, III, IV, Appendix I). This invcsti_;ator believes that the lm·:est section. of the stream serves as a stoppin5._ over area v1here ~he gray lin~ stop on their· migration, v1ai tin:; the subsidence of the water. The depth, conbined ·with the slow current rr:ake 3 these :pools ideal durinJ; the sprin:s runoff. As the water and the current continuo to s1s.cken, :c::rne.llGr and mr.H11er fisb. ar8 l l J ] I J J J J ] J J J ] ~-j ~- ] ] ] ] J able to -make the journey. ·Such a movement would be difficult earlier because just above the lower s~ction the character of the stream is such that there. is a complete absence of protected pools, resulting in turbulent and very strong currents. ( By the end of June, the fish.were.found to be still moving upstream, although by then individuals had already reached the ' uppermost portion. The fish in·the lower section were still moving. This was observed visually, and was verified by color tagging four fish in the first week of July.. The rapid changes in distribution of fish in the stream seen in Figures 4 and 5 took approximately one week. In July the upstream movement appeared slower than it had been in late June. At this time only a few smaller individuals were seen moving upstream. By the last week of July of both 1967 and 1968, l~rge grayling had virtually disappeared from the lower section, but considerable numbers of fish under 160 mm in length were still present. Once the fish reach the upper or middle sections of the stream· they tend to settle dovm in individual pools where they remain for the rest of the summer. Marked individuals have been observed living in the_same pools for as much as five -vveeks 7 but thes;e will .be discusseq. in a later section. During-mid-summer the distribution is uniform throughout the_ stream (Fig. 4), with the largest iridi viduals concentrated :1 J J J J J J J J J J J J ] J J J ] J 27 in the-upper half. The habitat in which the individuals ·settle down for the summer consists of clear water of 0-4 ppm turbidity and 97 to 100% oxygen saturation. The fish in the upper reaches stay near the bottom ( by day, and leave. the pools as darkness approaches. I was unable to see.what happened once they left the observed pools in the evening. I .suspect that a random movement upstreamand downstream-takes place, which would explain .the random distribution shown by Figure 6. In the lower Chatanika River grayling were reported moving upstream as late as the last week of September, when 850+ grayling were counted at a weir nmintained by the Alaska Department of Fish and Game. These fish were accompanying a school ofwhitefish (Gene Roguski, personal communi cation) • No data were obtained beyond September 15, but reports f'rom fishermen indicate a do\mstream movement at the end of the month, when fish were being taken in pools between stations 0 and 1, an area \vhere they had been absent in both spring and summer. At this time, the water was too low.for the grayling to be able to jump over the dam, and the ·only place they could have come from wa·s upstream. Other data that might· indicate a late fall or \'linter downstream migration' are the capture of six tagged grayling l J J J J J J J J J J J J J ] d J ] J 28 below the dam in the spring, coupled with reports of fisher- men taking many tagged fish at the dam·in early June. The summer movements described above appear to follow closely those.observed by Shetter (1937) while working with brook trout in Michigan. There, the brook trout remained .in their pools th:r~ough the summ~r, and migrated. 3/4 of a mile upstream on the average in autumn. ·During t.he winter there was a downstream migration of the bulk of the popula-. tion for as far as 18 miles. This was followed by a returnto the tagging locality on the following year. The grayling; of McManus Creek differed from the trout studied by Shetter only in that their period of. stability in a pool did not last through June, July, August and early September, but only through late July and August. l J J J "l J J ] J J J J ] ] J ] J J ] c] DISTRIBUTION, TERRITORIALITY, AND SOCIAL HIERARCHY OF THE GRAYLING WITHIN THE POOLS MATERIALS AND rvTE THODS In the summer of 196ft six pools were selected for extensive studies. Tlrvo we.re selected from each of the main sections previously described. The selections vvere based on the following criteria: a) pools vlhich were represent;ati ve of the section in which they wer~ located. b) ease of access. c) pools allowing_ observation· even during periods of relatively high water. d) availability·of observation posts which would not alarm the fish. The pools were designated as AA, AA 1 , BB, BB 1 , CC, a:nd cc 1 (Fig. 1). Pools AA, AA 1 , and CC were visited wee~ly or more oft.en after having been established; pool BB was visited biweekly. Pools BB 1 and CC 1 were visited three times, once e_ach in July, August, and September. Pools AA and CC were under observation from June 25 to Sept.ember 14, from 0'700 to 0200 on-the days of observation; the rest 1,..;ere observed from July. 22 to September 2. On each visit the observer approached with caution and 29 l l J J J J J ] J 9 ~J J J J ,-1 ~.J ] ] ] J J ~: 30 sat in a position allovling clear view of the pool, while still having a brushy barrier present as. camouflage •. From fifteen minutes to one half hour was allowed for the fish to adjust to the observer. Visual observations of the distribution of the fish, both vertical and horizontal, were made, activities that the fish were engaged in were described either in writtenform.or on tape, and whenever possible Super 8 movie films of th.e activities were .taken. At hourly int~rvals water temperatures v.rere taken. The physical characteristics of the six observation pools are tabulated in Table l. In Figures 8 to 12 the chatacter of the pools and the distribution of the fish within each are illustrated. To determine the areas occupied by individual fish in pools AA and CC, the fish's movements over a ten-minute period were plotted. · This was done by .scaling a paper· into ·1 inch equals l foot sections, and tracing all movements as they occurred during that period of time. The horizontal distribution of the fish was determined visually from above the surface of the water. The vertical distribution was determined with the aid of a glass bottom bucket. L..J L.J L...J CLJ c...J LJ L..J L......l Lj ~ L.:...J t......J LJ L...J TABLE I. CHARACTERISTICS OF SIX OBSERVED POOLS POOL DATE MAX. MAX. MAX. RATIO NO. STUDIED LENGTH WIDTH DEPTH OF BOTTOM COMPOSITION (m) (m) (m) % % % Rubble Gravel Sand AA .6/25-2.05 1.10 0.70 0 so 20 . 7/23 AA 1 6/25-12.50 6.00 l.SO 0 10 90 7/23 . BB-. 7/22- 9/2 9.00 3.00 l ~35 50 50 0 BB 1 7/22-- 9/2 45.00 6.00 l.SO 100 0 0 -cc 7/16- 9/1(: 7·50 3.00 0.90 30. 60 10 cc 1 . 7/16-9.00 3.00 0.90 10 so 3.-0 9/16 t:....J [._J LJ L..:_j ~ .. r··, ,.;~ -..:-:-.'~ :_,·)"?~ ,-.~~;-~,-~ '~:~~;:01:~~·;·~~t~~b~f:~'7~· NO. OF~ FISH SIZES OF PRESENT FISH PRESENT s 3 15+ 10+ 12 s...:1o (in mm) 75 to 1S5 75 to 200 120 to 350 100 to 380 125 to 250 120 to 230 \,..V p J J J J J J J J ] J J J J J J ] 32. RESULTS Horizontal and vertical.distribution of the fish ·---- In the six pools studied, and most of the other ~7 pools observed, the larger fish occupied positions close to t!).e bottom (Fig. 12), and the smaller on~s v1ere. ciistri- buted<iri the mid-depths. In the same pools the largest members were found in the.most forward positions in rela-. tion to the current (Figs. 7, 8, 9," 10, and 11), and the smaller ones trailed behind. When a pool Y.Tas observed in cross section, it was found that the largest fish occupied the deepest portion of the pool, often corresponding to the center, with smaller fish on either side, and the f.ish becoming progressively smaller as one neared the shore. The generalizations stated above held true v1hether the pools had a population of parr, adults, or a mixture of the two. The latter was the most common combination found. The nature of McManus Creek is such that the deepest portion of each pool usually is to be found immediately below the·upstream riffle. In this part of thepool the current is swiftest, and the only rocks deposited are of rubble and boulder size. Due to the uneven bottom, the large fish occupy positions immediately behind large rocks, \•There the current is not as strong~. ·Measurements taken J ] Fallen tree ] Rubble banks J J Rubble bars J J Vegetation J J Sand J J Direction of current ] Key to the follovring ·five figures. J J J J J l l J J J J J J J J J J J J J ,j J J J . 33 0 J . 2 · feet . Figure 7. Pool AA. Viev·Ted from 2bove c:md looking upstream. ] ] J J 1 d J J J J J J J ] J ] J J 34 ·+· E Figure 8. POOL BB. Viewed from above and looking dmvnstream. l 35 ~I J ] J J ., J J J ~o v ., ~o ., ~o ~o .., J .., J J J J L· •.'. ·,..., J g Figure 9. POOL BB 1 ~- J Viewed from above and looking downstream. J r ~·' tL· }-- J J J J J J J J ] J J J J J J J J J Figure 10. ~~:1,~l~t.,;•« ~o ., ~o POOL BB. ·viewed from above and looking downstream. 36 J J J J J J J J J J J J J ~] J ~J ] J ~? . J w ·+· 1 E Figure 11. Pool ee 1 • ~~~~~tf.i~TI~:~P:~§ ···· .. :(:t~(/·~~~e~r:{~t!~~ •• .• ~.~ .... ,. •·. (.~"t.,...;.~. l ·'-~:;..t ...... .\ •. 1;.. .•• , --~----~-... ·-·~·.,·1 ~%~;~~~~~t~l:~~~~~i~~~~ .\:;~. ·t: ( -=-''Po:"'\:·::{.,;· .. ·~::.: {),-";·t ..,;.. (..~ .. ~~ .. ~r·f."\ llltll MONTAN·~~CRt'if.'<~~~,--~;,;~~ .1:-l:~:~. ~: i.· r-<·!":~·'!·:A-~ • :· . .1. • . ~<·t-"7··~ •:A ....... !:' •• ·:.,.?· .,_·. ·-·:;-• ..J·£.· •• t. •. y· ,._.-J .. · ·>. ·: . ."-\•· .. ,... ... ·"'· ~. -~ ... ·_t.: •.•• ·.1'-·.l.· Viewed from above and looking downstream. 37 J J ] J I J J J J J J J J ] ] J d ] ] J Figure 12. Vertical distribution of grayling in a pool. l J J ] J J ] J J J J J J J J d ] J J 38 l J J ] J ] J J J J J J J J J ] J 39 in the summer of 1967 indicated that the velocity near the bottom, at the head of the pool, was si~ilar t'o the velocity _at mid,.-depth further back in the pools, where mo.st of the parr were found. Exceptions to the distribution stated above occurred: ( . a) among adult fish which had been previously tagged. b) in long narrow pools in turbulent stretches of stream. c) in standing pools with little or no current. d) in all the pools observed after September 2. Adults carrying peduncle tags were seen in three cases, and one of these 1r.ras in pool CC, where it was observed f'rom June 25 to July 28. In all three .cases the tagged fish were the largest fish in the pool, and yet occupied a position behind a slightly smaller fish. The physical appearance of the tagged fish was poor. Aside from the swollen peduncle and tail, their bellies were withdrawn. The index of condition for two of the fish was determined by using the equation (lagler, 1956): K = K. = w = L = w index of condition \>reight in grams · length in millimeters J J J ] J J ] J 3 J J J J ] J J J ] K for the two tg~ged fish vvas 56.7 x 1o-7 and 58.9 x ·1o-7 , as compared to 109.2 x 10-7 for healthy fish of the same size, collected in the same area and during the same week. In the situation involving narrow pools in turbulent stretches of stream, the horizontal distribution also varied from the norm. These pools occurred primarily in the upper portions of the stream. In these the larger f.ish were typically found in the d~epest portion of the pool, which usually corresponded to a location half-vvay down the length of the pool. The parr vv.ere l.ocated both ahead and behind the larger fish (Fig. 13). The lack of typical distribution was also observed in the pool below the diversion dam (Fig. 14), in pool AA 1 , and at the junctiQn of Homestake and Charity Creeks at the headwaters .of Faith Creek. All three of these.pools were characterized by large size, deep water, and lack of a strong current • . ri'he pool below the diversion dam (Fig. 14) was the. deepest point in the drainage. \tlhen the water level dropped in late June, the pool had a considerable portion of standing water. As.soon as the water dropped and the current diminished, the fish, which previously had been lying close to the bottom, began milling around in schools. The movement of the schools was in a counter-clockt...;ise J J J ] J J J J J J J J J J J J J J 4l • til rl 0 0 p.. ~ -0 H H ro ~ ~ •rl ..c: til ·rl Ct-t Ct-t 0 s:: 0· •rl "+> ~ ...0 ·rl H +' til •rl ~ [._J L...: L.J r :· · 1 L.,..J.......J L_j [_J LJ LJ L..J c....J L..:J ,. r·1 '\..-..,;.J L-D ~ L..J L..J J. ..-.-., DAM I -(':-."·;·. r,07'•,....,.- /',,l\.(:l• •• ·.·>.·.~·1'~··, ;-:-,.. .... I< • • ·.)·'\"· o,j •• -~'£~"·,:.:-._ ' /' .. ·""''""'' -... •. .. ~~~·);'':..~~ I ·~.-?·;.. , -----. , ·-.~r.,,z;. /' ., .. ,t: o I ~ 'T~ .. :..t · 1 i1Yf / "" )t~i~ _.·;ft: , J~¥~{~ • 4.. . .l~:t''' .' "'. . . ~~~.;.~: .. >~7 I ~~~ .. •'7 ... ,~·1 I f'; t~ ~~d0 I / .-fft I >:·I \ ~.':i-:.; :,J.··~I ......___,.. ~ •• :"!::~·.: I f.~:··';.\ -~ .• v. ·,: !Y 1!~11 0~ . li· .... ··, . 11',)1._;.5 ..... l·'·..j.~ ... : ... :,· ..... _;!r , r-·. ·.~! . > . .;.:~.: '4/ I ,., .. \ ... ·,..·-t'' , ~ ·. '<:\ .... ' ..... L.. 7 ·j' '.::.::":~ !',"':·': :.!,!.;:~ ,, l·:v:.· "!'"·.,,.:.,.. ~-..;·v I 'y•oj' !~4•' ,.f:, "''!.'' ', f ~·~/~~~· • . :":·.~·~Y.~~?;.t~~ \ ·, i 1~:!·,::11 ~~·: .... :-.;~.~:~·~, t·;..:f .. ! tv.-;..·.~1 ] I r.,:r_;_':-·:r::r=.;,..:,:; . r:~ I ·~\:W.j\ 1 ·1 1 ~\:t;:;;w~~·-~ k·, ~·-~~~~~ I · .,;: .l · iA'~"-: .. -... ~·"·v.J.'... £.~~ -1-~ ·>f-;.,1 ~ y r .. • ·"'•v •>''·· ., {'"" • •t-:o .. M-'<:"t , ::lj 1.:·.·~·;..:·.~-'"·>·':>' '•>1·:"'··.:.·:·~·:::-. ~·· '~ f I· -~·.1,;. ,..: • • ·~o-.i,· :,... . • . •.;,.' •• "'.' ?. ~~· . •ol'\,..l •• ,. •. t--·~.·,.._·"'·'tL·v ...... ~ . --Movements of the fish in the pocil. ---·-Directiorl of the current. Figure 14. Movements of fish in "Che pool below the dam·. Solid lines indicate motion of the water, the dashed lirie shows movement of the fish. -, I .· I . I I I L.J LJ [_j -t> (\), J J J J J J J J J J J J J ] J J 43 direction, the direction of travel vias counter to that of the slight clockwise current present.-The schools had the larger fish in the forefront and center of the mass, with . . ) i smaller· ones at t;he periphery and trailing behind. In early July, when schools of round whi t.efish ( CorP~us, _cyli:r:dr~s) entered the pool, the grayling joined these school.s .and moved ·together as a group. Pool AA 1 , the only one of t'.he three pools that v-ras vlithin the study area, was also a large pool v-rith very slow current. Here the fish were wandering in a figure eight pattern. The best illustration of the conditions under vihich the breakdovrtJ-of the horizontal distribution occurred was the pool at the junction of Homest,ake · and Charity Cre.eks. Up until July 20, this pool was approximately 30 feet long and 12 feet wide. The distribution of 11 fi$h ·. inhabiting it followed closely the pattern illustr~ted in Figur~ ll for cc 1 • After July 20 a reservoir was created by an earthen dam constructed 50 meters below the pool. After the dam was built a reservoir approximA-tely It meters deep and inundating an area about 50 meters by 30 meters resulted. As the vmter began to rise, the fish left their previous positions and began moving around the pool in a school for the remainder of the summer. Because of J J J J J ] J J J J J J J J J j J J J the continuous movement of the school, the large size of the reservoir, and the lack of a suitable vantage point, observations. of the ordering of the school could not be made. One thing was apparent however, the number of indi~iduals in the school was constantly increasing through- out the summer. Whether the additions came through the discharge pipes, or moved down from the two tributaries .. was· not determined. On September 14, pools AA, AA 1 , ee, and eel were observed for the last time. AA and AA 1 were devoid of fish, ee and ee 1 still contained fish, but the composition had changed. Both pools contained only adult fish, arid the horizontal distribution appeared random, with all of the· fish staying close to the bottom. Evidences of changes in composition first became apparent on September 2, '"hen both. in ee and BB, fish which had been color tagged on ·July 22 started disappearingfor the first time in 42 days. Feeding range and feedigg center Using Burt's (1943) definition of home· range as "the · . area, usually around a home site, over \V"hi ch t.he animal normally travels in search of food," then the grayling observed from June 20 to September 2 can be said to have a home range~. However, this range will be referred to as a· feeding range since all movements considered here were l ] J J J J J J ] J J J ] J J j J J J t> ~~ f 45 fe·eding movements •. ·All of the fish l'Tandered from a -range center, but they always began their movements from it and returned to it. 'rhe presence of the feeding range and feeding center is illustrated in Figure 15 ~-These are two tracings of the inovernents of two fish, Figu~e 15a of fish #1 in pool AA, and .Figure 15b of parr #4 in pool CC, over a ten-minute period·. Eighteen such tracings \vere made in pools AA an(l CC •· Of the lS, only S included only feeding movements, ·and these latter ones were the only ones that could be used in relation to feeding ranges. The distances traveled are tabulated in Table II. From the table it was calculated that the larger fish seldom moved more than 60 em in any one directiqh in search of f0od, only 6.S'fo of the movement,s exceeded ?0 em,. and 4.S%were greater than one meter. The parr, hmvever, designated as Group B in Table II, had on the average larger home ranges, as 15.9'fo of all movements exceeded 70 em, but only 5.2% exceeded one meter. The range centers of individual fish v-rere very definite; individuals could often be recognized by their association with a pebble no larger-than four centimeters .in diameter. After excursions mvay ··from the range center the fish would return and occupy exactly the same spot. These sites were semi-permanent, that is they remained J J J J J J J J J J J J J J J J J J 30.cm a)Movements of dominant fish in pool·AA. b)Movements of fish.#4 in pool CC. Figure 15. Tracings of the movements of two fish over ten minute periods. 46 [__;L..JL.JLLJL_j[._JL._J(.__JL.Jc:..JJL..:JL..Ji__jl_jL...JL..JL_.JL...:..J~ TABLE II. MOVEMENTS OF GRAYLING OVER TEN-MINUTE PERIODS GROUP A DATE 7-9 7-16 7-23 Mathema.ti- cal mean for Group A - GROUP B 7-9 7-9 7-16 .. 7-16 7-23 Mathemati- cal mean for Group B POOL NO. BB 1 cc cc BB 1 AA cc AA cc TOTAL FORWARD 5 7 9 7.00 23 14 27 31 30 25 .. 00 ·o-30 . 30-6o 60-90 90+ em em em em 4 0 0 l 5 l 0 0 4 4 l 0 4.25 1.66 0.33 0.33 7 14 2 0 3 3 4 4 6 15 6 0 17 6 6 2 19 6 2 3 . 10.4 8.8 4.0 1.8 TOTAL BACKWARD 44 38 42 41.33 52 45 4""1 48 35 44.2 0-30 30-60 60~90 . 90+ em em em em 1'6 12 0 0 28 8 0 2 16 20 2 4 15.00 7.66 0.66 2.00 32 18 0 2 21 17 7 0 25 7 ~-5 30 18 0 0 20 7 6 2 25.6 13.4 3.4 1.80 .;::- -,J J J J J J J ] J J J J J ] ,J J j ] J J the same throughout the summer for as long as the composition of the fish in the pool did not change. If a fish was removed from a pool the rest of the fish behind him would move up and occupy different feeding ranges. Territories and Hierarchies --. -oo;--.. 1'he individ:ual grayling in the pools-exhibited restricted movements over·a small area, theivr feeding range. An area slightly larger than. the feeding range was defended against other members of the species whenever the latter entered it, thus exhibiting definite territoriality. The first time· that the pools were visited on t.Tune 25 the territories had already been established and the procedure t!J.a~ the fish had used in distributing themselves was not seen. However, due to the fishes' peculiar fright behavior, some insight on how the original spatial-distribu- tion might have taken place can be gained. Whenever the pools were disturbed, either by someone walking along the bank or seining, the fish scurried at fir.st to the deepest portion of the pool and if the distur,;_ · banc.e persisted they moved to the downstream end of the pool. Slowly, after the disturbanc·e subsided, the smaller parr moved to the head of the pool in the vicinity of the position formerly held by the largest fishes. These parr would return mostly in small groups of similar size, ·the groups of the l J J J u J J J J J J J ] ] J J J ] 49 smallest fish coming first, until .appro~imately 15 minutes later when the largest fish returned. -As these groups returned· to-the head of the pool, each returning individual began a display \vi th its closest neighbor which had already settled in the area, seemingly trying to displace the latter and usually succeeding. Each fish did .not display with every fish smalle,r than itself, but only·. l . . with the one or t;.vo slightly smaller than itself, and only ' to regain the position held in the pool before the disturb..,. ance. As a fish was displaced it moved downstream in the pool, and as it did so, so did every other smaller fish behind it, maintaining fixed distances between the members of the group. For pool CC (Fig. 10) the distances between fish were: Fis.h #1 and #2 -approximately 25 em Fish //2 and #3 -approximately 25 em Fish #3 and #4 -approximately 50 to 60 em In pool AA (Fig. 7) dealing with smaller fish, the distances between fish were as follows: Fish #1-and #2 40 em Fish #3, #4, and #5 -20 to 25 em Smalle~t fish -8 to 15 em J J J J J J J '· J J J J J ] J J j J J ] 50 After the pools were disturbed the returning fish began to display with each other~ These displays, which are called challenge displays, were always initiated in the same manner. 1. This began with the "invader" moving up to the "defender" until parallel with him. 2. · Both fish then lay very still vii th dorsal fins folded across their back. 3. . i . . Then slowly the -invader would begin drifting toward the other until their bodies were less than one centimeter apart. 4. Then one, u~ually the invader, moved forward until 14 or 15 em away and assumed an arched position so 'that the inside of the arch was presented to the head.of the other fish. 5. In this position the arched fish, the invader, moved backwards and as it neared the defender, the invader began sinking until beneath the defender. 6. Once beneath,it began rising in the water column and as the two fish drew close, the defender would begin drifting backwards (Fig. 16). At this point one of two things would happen: 1. The r~treating fish kept on retreating until it was out ofthe area. 2. The retreating fish moved around the· rising fish until parallel and then the pattern of events was repeated with the protagonists reversing, their role. l J J J J J ] J J J J J J J J J J J J Figure 16. Steps in establishing dominance in a set. a. Invader approaches dominant defender. b. and c. Irwader displays and sinks below defender. d. Defender begins display. ··~ .. ·:: J J J J 'l J J J J J J J J J J J j J J J 51 ' J J J J J J J J J J J J J J J J J J 52 The display as described above. happened qu;i.ckly, .. and v~as observed in its entirety only five times when com;ii tions · allowed the use of a glass bottom bucket. Many similar displays had been seen from the surface, but, due· to the turbulence and surface reflection, some of the detailswere ( missed.· Thirty~seven·observations·made from·above the-sur- face all.share a striking similarity to the display described above. However, ·it should be note<i· that the description above is one of the most complete seen. In most cases the do~i­ nance between fish vvas determined by step 3 ·. ·The most intense displays occurred among parr 150 to 200 mm long. Among the largest fish the displays very seldom reached step 3, usually terminating a,t step 2. In pool CC I never saw a display between fishes #1 and #2 go beyond step 1, as fish #2, which was wearing a peduncle tag, was prompt in retreating. The tag-bearing fish, although larger, was found occupying a less dominant position, moved less than any other fish, was frightened more easily than any other, and ~ecovered from fright much slower, so that it always was the last fish .to return. The great majority of displays in my notes appear as follows: July 13, Pool CC, 1700 "Involving fishes #3 and f/4. The two fish moved until parallel to each other, then the aggressor moved forward five inches and assumed a scimitar stance, and began drifting tm-vard the defender, the latter.drifted back with the current and left the terri tory. • • " J J J J 0 J J J J J J J J ] ] ~J J J J 53 In all of the pools, if the lead fish moved either forward or ba(!kward, the rest of the fish responded by doing the same as a group. ·The critical distance that the lead fish had ·to move before the rest of the group followed was not deter_mined. Nine times in pool ·AA, and five times in pool cc· the lead fish rushed up to the riffles and spent several minutes there. Slowly the fish occupying· '#2 position moved up until they occ~p~ed the spot previously occupied by #1. On #l's return the fish did not move back to their old places until =/11 displayed mildly with #2, at which time all of the fish inoved back to their old posi tion.s. In pool CC it first became apparent t-hat the positions occupied by #1 and #2 were, for the same reason, favored ·by the other fish in the pool. 'Whenever one of· the t'\'VO moved- out of the pool, their exact positions were taken ove·r by fishes #3 and #4. This type of behavior was especially 1rmll demonstrated in this pool when fish #2 vanished, apparently taken by a fisherman on July 28. Upon his disappearance fishes #3 and #4 moved up to #2' s position~ The.re a series· · of interactions in the_form of displays bet~een the two fish took place, until #4 remained and #3 moved approximately 30 em back. A week later a fish· approximately 250 mm in length had moved into the pool and occupied exactly the same spot. previously held by #2, and #3 and #4 moved back to their former positions. J l J J J J ] J J J J J ] ] ] J ] J J 54 Similar behavior had been exhibited in several of the . . pools used for the collection of fish. In these, within 24 hours of the removal of the la:rger fish, the inhabitants had reorganized themselves by Il'loving up and occupying the exact spot of the removed fish. Once the territories v.1ere established, the fish still _occasionally interacted, as smaller fi$h a})peared to b~ ~ . constantly attempting to move forward in the pools •. Competi- tion for position was especially keen among parr. occupying the middle of the ordering-, and considerable energy was expended in maintaining and defending the position in the pool against fishes of similar sizes. The larger individuals V'rere especially tolerant of fish much smaller than themselves to be present ahead of them. Thus, in pool AA 1 there was a parr, #3,.which occupied a. position upstream and north of fish #1 (Fig. 7). This situ~ ation was allowed as long as #3 did not move toward the center of the current. Whenever #3 moved across line I-I (Fig; 7), which denoted the end of a sand bar, he \vas subject to attack by #1. If #3 moved too far back he \-vas promptly .attacked by #4. In pool CC there was a fish approximately 130 mm long between fishes #1 and #2, but off to the south. This fish v.1as not allowed to come tmvard the center by #2, but otherwise was.not molested except by #'s 3·and 4 who regularly tried to chase the fish out of the area at least l J J J 'I J J '1 J J J J J ] ] J ] ] J J once per observation period. In the n chases'', 113 or. #4 would rush toward the smaller fish, which fled to the 55 head of the pool only to return slowly. a fe\'IT minutes later. During the defen~e of territories by the various fish, the following types of ·display~ took place: The invading fish would move to a position ahead of a larger fish, the latter would make guick rushes to the .. invader's flanks, when 5 or 6 em apart, and 4 to 5 em up- stream of the invader. Then the defender would slide or drift backward toward the other fish at a position 30° to the current. ·Most of the time, 33 out of 42 observed displays, the invader l"muld drift backrt!ards as the defender got to ,,.Ji thin 3 em of the invader's snout (Fig. 17) • Tvm variations of the above display have been observed. In one the defender simply rushed up to the invader,' s ·side and the latter quickly lef't the area, swimming \"lith the current and the defender in pursuit. Actual contact \tfas never.made. In the second variation, the invader did not leave the forward position when rushed, but moved instead to.one side of the pool. This behavior was observed only five times. Each time, as the invader moved toward the side of the pool,. the defender returned to hi~ usual position. A few seconds later, the defender followed with a·new rush as before, to which the invader responded by moving .to some other position or by going back to· the lmver end of the pool. After the l J J J J J ] J J J J J ] ] J J J J Figure 17. Steps in territorial defense. a. Invader assumes dominant position. b. Defender rushes invader. - c. Defender displays to invader. J J J J J J J J J J J J ] ] J ~ ] J. n ~:~ ~·· ~£,~ ] 56 i" . l l J J J J J J J J J J ] J J tJ J J J 57 second charge, a51 happened in three of the five situat:i,ons, the defender vmuld rush and pursue the intruder, l).arrassing . I him continuously until he left the territory. Of the three incidents, twice the invader left the terri tory on the .third rush and once on the first rush. ( . . Fish of similar sizeseldom interacted, and the few times that they did, each individual returned to its range center l after very mild di~plays. On July 16 I recorded a defense of an area by more than one fish. This was the first time that I had observed this, but similar behavior was later observed twice at pool CC. In the incident at pool AA, fish #5 moved next to #4 and was promptly rushed by #4. '#5 seemingly. ignored #4 and moved to a position betv-reen #1 and #4. Both #1 and #4 rushed #5's flanks, stopping approximately 1 em from actual contact. After #1 and #4 joined the melee, #5 swam back to its original position. Figure 18 shows the movement of #4 after another larger .. fish moved to a position approximately 60 em ahead· of #4. Note the increase in area covered during the· ten minutes, and the four dashes toward the intruder, which are labeled I,II,III, and IV. These were of decreasing intensity and took place only during the first two minutes of intrusion. In the fall the hierarchial ordering appears to break down completely· (Fig. lOb). On September 16, the breaking down of the hierarchy was very noticeable at pool CC, -.;.;here J l J J I J J J J J J J J J J J J J J L __________ c --------- H' .; o-' "(' • 0. J. ·'·1..:> 1_,_ __ e J.o. Movements cf fish in torrit~rial defense.· .t..rrow indicates direction of current. See text for discussion. J J J J J J J J J J J J J J J J ] J J ' r 'V ~; 59 adults and sub-adults were intermixed bor.izon.t<ttly, a:rid )ill occupied positions close to the bottom. No displays o!-' any type were· noted over a seve:q. hour pe:r-iod, altbough t1tlice smaller grayling moved rapidly back in the pool upon the a:gproach of their .counterparts. l J J ] J J J J J J J J J J J ] J J ] 6 !') \..1 . DISCUSSION In the distribution of living organisms it is found that many animals tend to aggregate, mostly for mutual benefit (Al:Lee et al., 1949). The-se aggregations,.particu- larly among vertebrates, are brought about primarily t}:;rro-q.gh three'major plrinciples o:f behavior---territoriality, hierarchy, ang. home range. Most animals display just one of these prin- ciples, but subtle traces of the other two can of.ten be found. Although many kinds of territoriality are recognized (Allee et.al., 1949, p. 412), the most. commonly .described is the mating territory, in which the nest.is defended against all intruders·by one or both members of an established pair. In fishes this was first d~scribed by Noble and Curtis (1939) '(~> ~~:. . .,. ..... , -. in the cychlid Hemichromis bimaculatus, and in the poeciliid Xiph<2£llorus helleri by Noble. ( 1939). Territoriality in non-spawning individuals has been described by Greenberg (1947) among immature and.adult green sunfish (J-. .. ~12omis cyanellus). In this species the territories established were defended by the residents against subordin- ates but not against dominant individuals. The concept of home range in fishes has been used by Gerking· ( 1950, 1953) in discussing the stabi1i ty of f'ish populations in streams, and later by Nevnnan (1956) in the study of interspecific. competition bet"\11/een two trout. Although all of the studies mentioned above use Burt's ' _.=j --, ___;. -, -'-" -" 61 (1943 )· definition of· home range, none of the 1rwrks on fish / . used the definition as the authors had defined it. Gerking (1950, 1953) .and Newman '(1956) when discussing the home range and home. areas of salmonids, interpreted the home range·as being the pools in which tJ:ie fish had been tagged. Some of ·their pools were in excess of 300 feet long. When discussing movements in an out of the home range the authors interpreted it as meaning movement from pool t.'o pool, and this was tes-ted by the· tag and recapture method. However, l\'filler (1957) showed that the home pool was more than a home range for some trout, because some trout were born, grew, reproduced and died 'lrlithin one pool, and different parts -of the poor were used for differ- ent life processes. Further studies might indicate-that the home range in the pools, using Burt's definition, might be considerably smaller than previously thought. The present study is unique in that a home pool was observed as·· a unit, and its inhabitants 'i."fere observed indi vidu- ally and in relation to each other under natural conditions. The data indicate that the Arctic Grayling in Mcl'vTanus Creek during the summer indee4 have.true home ranges as·defined by Burt (1943). In these "home ranges", which here are discussed as feeding ranges, there is a range center 'l.rlhere most of the time is spent, and from 'l.·.fhich all forays, whether 'feeding or othe:ewise, begin and end.· These fe_eding ranges varied little _ v'lith the size of the fish, although parr appeared to make slightly longer feeding forays than adults. __; -, ~-, ~ --., ~ ~ -• 62 . Gerking (1953) defined defense of territory p.s 11t.he aggressive response of an animal for the_protection of an area from invasion." Following the above definition, then, the arctic grayling in,the stream do exhibit territorial defense. The territory defended by these fish tended to be slightly larger on the upst!'eam si~e than the feeding rQ,nge, and· only the area ahead of the fis-h was defended.. The terr-itorial defense movements differed from feeding movements in that they were more intensive (i.e., quicker movements), ··and often covered a distance twice .as long as the feedipg movements. In the defense of territory the grayling followed through what appeared to be a ritualistic display. They bent their body in a scimitar shape while drifting toward their opponent. A similar display is described for PlatYP.oecilus maculatus by Braddock ( 1945), "~1vho interprets the bent body as a challenge movement·. Fabricius and Gustafson (1955) briefly described the aggressive behavior of the European grayling ThYmallu~ thy!!!:allus. T. !:.hY!!!allus apparently is much more aggressive, its activities involying considerable physi.cal contact and. resulting in nipping at each other's tails and flanks. Aggressiye behavior also involved violent vibrations of the body and erection of the dorsal fin, none of which occurred in :McManus Creek during the period studied. The dorsal fin --, ., ... __ ,.i 4 -" --' .... ..... .Jl -' :~ ii' ~-~- ff. ~ ~· t 63 o.f the arctic grayling ·is larger than the ones on the. Eu-ropean ::;pecies, and the musculature controlling the move-:- rhents of the dorsal rays, as revealed by dissection, appears poorly developed. This, coupled with the creek's fast and . turbulent current, would make the use o.f the dorsal . .fin .as ( a display.organ disadvantageous. Due to its size, itwould offer, when extended, a co:nsiderable surface area for the current to act upon. Hence, gainsvobtained by using such an organ for displays would probably be offset by the energy utilized in fighting the current. Perhaps the fish in the slower Alaskan streams and lakes mayemploy the dorsal fin to a larger extent.. The lack o.f utilization of such an obvious organ as the large fin of the arctic grayling pre- sents a puzzle. Its size and coloration would lead one to speculate that it ~tTOuld have great adaptive value in the life .of the grayling, and yet none was found. It is unfortunate that the spawning behavior of the fish was not observed, because possibly the answer to the fin's significance might be found there. When the fish established their distribution in the pools, it was found that the distribution v1as attained through a series of aggressive displays, in which one fish became the dominant and occupied the forward position in the pool. In .each set of displays a set of dominant-subordinate relation- v~ ~).:-::-__ ships -wer-e" formed. The subordinate fish expressed his submission.by backing off.slowly 5 to 10 em, and then fleeing ~ -, .._, .., jj "_d ~ .... -~ __. --' __. _. 61*' with the .current and re--establishing a feeding or territorial center 25 to 50 em dm·mstream. The establishment of sets of dominant-subordinate relationships automatically resulted_in the establishment of a hierarchy. This hierarchy v.ras not a true straight line ( . ·. hierarchy~ since many of the smaller fish were allmved between and. ·around more . dominant fish (Fig. 19) . It resembled more . l clQsely a pseudo-straight line hierarchy \-There the positions of one or more individuals were undetermined. The undeter- mined individuals consisted primarily of the smallest occupants of the pools which were often allowed in positions close to members several times larger than themselves. A hypothetical hierarchial ordering . of the fishes of a pool is shovm in Figure 19. Fishes setting up._ territories in· a stream are faced with environmental factm:·s niuch different from pond-living fishes or fishes held in tanks. The primary difference is the presence of a strong current. Most lotic organisms, and fishes are no exception, exhi_b_i t rheotaxis ( Fraenkel and Gunn, 1961). That is; the fish turn so as to-face the current and spend most of their time and. considerable energy swiniming just strongly enough to keep their location iri the stream. Rheotaxis \'las brilliantly demonstrated by Lyons'· experiments in 1904 which are sumrriarized in Fraenkel and Gunn (1961). His fish, \'Thenever placed in standing water, displayed milling behavior, while in running water or·simulated running water.they.always '-1 IIi Ill Ill 1'1! I~ . i~ ~ ~I w~ I i ~ j I' -~ • I . ~ 'i I I I I • lJ ~ . 0 -- : _.; --., --:> -, _..;J _. J -"' _. C7 <=:::::><t j' c:;;' ~ l2 Tog ~ 13 &;? f) <:::::>~ r - (:f) ~ l Indi-viduals of these groups were tolerated by fish2 and 3 as long as they were not situated in o direct line between the two fish ' Distribution of smaller ones often breaks down Figure 19. Hypothetical hierarchial ordering of fish •. 65 t~ ~ ~ !l • ~ ij --' -" -, --' ___. _. 66 oriented themselves against the curr.ent. Grayling. d-isplayed this also. Specimens removed from the Chena River in October 1968 and placed in standing tanks oriented .themselves at random, and when disturbed exhibited milling reactions. The same fishes moved to a Liv:lng Strearri tank immediat~ly ' oriented. themselves facing the. current and remained in this position for the month that they vtere held. Increasing or slowing the current did not affect their orientation. In the field, such observations were complemented when the grayling in the·reservoir at Homestake and Charity Creeks.broke up their pool ordering as soon as the \vater in the reservoir began rising and the current all but disappeared. It is interesting to note that schools formed in standing pools, such as the one below the diversion dam, show a similar within-school distribution as the distribution of the fish in the pool, the primary difference being that the distance bet1veen the fishes was much reduced. That is, the largest fish are found in the front and center of the schools, arid. the smaller ones occupy the outside and rear. Due to the grayling's rheotactic orientation and their .feeding habits (see section on feeding behavior), the gray- ling's territories differ in shape from other territories described for fishes inhabiting len tic waters. In the grayling'_s territories the center was located at the downstream portion of the guarded area. All of the defense movements take place --. ~ __, -_, ; " __. _. --" ___. --" 67 upstream or to the side of this position. If one,~o.o.ks at the distribution of the fish and the shape of their territories within a hypothetical pool, the result would be sets of semi-lunar territories, one behind· the other, broken down :j.nto smaller but more numerous such territories as one moves do:vmstream (Fig. 20). I suspect that if more stream inhabiting fishes were closely observed in their natural 1 habitats, similar within---pool distribution and possible hierarchies would be found. Greenberg (1947), Gerking (1953) and Newman (1956) describe the hierarchies displayed by the fish that they studied as a nip-right orderin.g. The dominant fish in nip-right hierarchies asserted their dominance.by nipping the flanks of subordinates. The more dominant individual was noted, by Greenberg (1947) and Braddock (1945) ~ to be nipping more frequently· ~han the others. vli thin the frame~ work of their fishes' hierarchies there "l:vas a steady. dovmi'Vard progression in the amount of nipping-done by fishes occupying ·the several ranks. In grayling the nip-right type of.hier~ archy 1·vas not present, since physical contact between display- ing individuals was never made. Contrary to what Greenberg and Braddock observed in their studies, with the grayling there v-Tas an upward progression in the· number of displays occurring by the fishes occupying the several ranks:. Thus, . ~ . the more dominant fishes very seldom displayed. Seemingly, • r ,.c" ·r-, --~J·" .i'!.vt.. >+ .-v~c..t..~ + papUBJBp JC S~1..1Upunoq 8~BW1XO..IddB 8~BD1PU1 S8U11 TBD1~d111a· "100~ B U1q~1M U01~nq1Jqs1p TBD1~aq~od£H "0~ aJn~1d ,, ---., ---., -~ 3 ' ~ ~ _; __) _J 69 the size of the dominant indi victuals was eriough to in'hibit any challenge motivation from sv.bordinate members of the pool. The displays observed during the formc;1tion of the distribution appeared toindicate that dominance was achieved primarily by the more aggressive and biologic~lly better fit member of the pair, Vlhich usually was the larger member. This explains the secondary position occupied by the tagged fishes, even though they were the largest fishes in the pools. Hoar (1953) found tagged coho smolts (Oncorhynchus · kisutch) also occupying subordinate ranks and in poor condi----- tion. He attributed this to their being kept from feeding by the healthier ·individuals. The breaking dovm of the hierarchy of the grayling in the fall appears similar to situations described by Hoar (1953) .. He found that coho smolts were characterized by aggressive behavior during the seasons when they occupied particular locations. The smolts lost their aggressiveness during periods nearing migration time. The grayling lost their aggressiveness, and thus disrupted the hierarchies in the pools in September. This vv-as a period vlhen considerable i . movement v-ms taking place, as shovm by the complete changes . (~ ~ !~~ h\ [!'• t F ti l in composition in the pools. --, .., ...., =..il -, __) --, .--"' FOOD HABITS I•llE'I'HODS OF STUDY Grayling were collected weekly in the upper and middle sections of the stream from July.l to September 16, 196$, fordetermination. of food habits. During each collection period, the relative abundance of food organisms present in the stream was determined by collecting the benthic organ- isms present in three square feet ~f riffles by means of a · standard square foot Surber net. No attempts were made to determine the abundance of terrestrial or pool-inhabiting benthic organisms. The organisms collected were prese~ved in 70cfo ethyl alcohol and later identified, counted, and sorted into the various taxonomic groups (Table 1, Appendix II). Their abundance for each half of the month was determined (Figs. 21 and 22). A total of 6$ stomachs was collected, none of which was totally empty. The fish -v-vere taken by hook· and line, ·and it appeared that no regurgitation of food took pl~ce. While the fish were :fresh, their stomachs were removed and placed in vials containing 70cfo ethyl .alcohol. In the laboratory each stomach was treated as a unit. The contents were removed and the total volume determined .• Then the contents were sorted under a dissecting microscope into various taxonomic groups. The numbers present from each --" -. .--" " ~ --" _j ' ,d --" I *- ~~- ~: ,. ~ t r:· fi ~" t' ~1_· ~; 50~- 40 q 30 i-\ \ \ \ c; \ \ \ \ \ o 0 Ephemeroptero o-----··-0 Tendipedidoe X )( Simuliidoe X-------X Plecoplero An11ellido H: 20 \ \ "" "' \ A. \ ., \ ::e \ ~ z \ \ "' .\.'> 10 \ 1!!1: \ 0 \ ... ~\ 0 I)( w ~ ~ _,.-0 :;l 5-z ---' ·-.., ________ ;'"')( ----v --.:::, ---7" '0 x-·----------7'--------~----. . ·---~-------:.._! ____ . ____ ·_· -1---~------:-; -·· .. --~· -~-.... ~ -·-·---- B A 8 A J 1J l y AUQUSf S E ? T. Figure 21. Absolute abundance of organisms in the stream~ A = first half, B = second half of each r.Gontl:1.9 71 _/: ~ ' " -, _, .. E .. c 0 Cl ~ 0 -0 2 0 ~ -" _ _.; ___. --" _.;.; 60 50 .. ,. ... .,. ... ,. .... 0 ..... / I ....... ?2" ,....o 40 ,__ // o---~ ~phe.me,:~ptarO 30 20 10 ~ I \ / \ -I ' I \ I \ I \ / \ / \ / \ / ' ,/ \ / \ / \ // \ / \ / \ / \ // \ / b" I I o-------o r~ndipedidt1~ X:---~ Simu!iidce X·------·---·X Plecoprera -----Ann~!llida _,.,.., _ _,, .... \x -x---------\ ,...-..::-.:::-=:-=:::.:::::___________ - .x----_,. ~~""'"' -,.... x--" ~ ____ l ___ L ________ L __ _ A J U t Y B AA\JGUSTIJ S .E -~ T Figure 22._ RB1ative abundance of-organisms in the st,:r~8a.m ~ A = fi:cst hal.f, E :::: f3econd haJ.f of each month. -., --, " ~ "' ~. -' --' -; '""' --"' _; =' \_" \! ,. i::-, £: 73 group and the volumes were determined. Vqlumes were ·deter- mined by W'ater displacement using a calibrated . c~ntrifuge tube. For small items the 0.2 ml marks were used and estima- tions were made. ·For food items too small to be measured, only their presence was recorded. ( The frequency of occurrence, volumetric method, and numerical method of analyzing stomach contents were all ' "· employed because it was felt that one method alone would not give a true estimate of the importance of the food items. Consideration of only the numbers of the various food items will not give a good picture .of the importance because of the large differences in size of individuals. yolumetric measurements are biased ·in favor of larger items because they take longer to digest. Frequencies, on the other hand, tend to be biased in favor of smaller items (Windell, 1968). From the three methods of stomach analysis, an attempt to determine the relative importance of the aquatic organisms . utilized by the fish as food items for the summer '\'lias .made • . On standard graph paper the frequency of occurrence was plotted on the Y axis, on the po.si ti ve X axis the· mean percent of organisms per stomach vms plotted, and on the negative X axis the mean percent· volume vms plotted by breaking down the total volume of the recognizable food items so that they occupied a segment equivalent in length to the segments of the ...... _;, ' ~ ~ _j -' 74 other two axis. In this manner, for each food item a triangular area was enclosed, and the.ratios of the areas formed by these triangles vmre used as relative indice·s .·of importance (Fig. 29) • --, '""I -., -' " -- --., •• JI ---, -" .J ~ ~'. t ~ " ~: lr ~ ~-:.:. t· t ,,, ~ H· ~~-· r,~. l ~;: u J K ~ ~ ·~, r i . 1'~. ' ' Ill __ · t ·~ 1~ . ' ' I . -•... ·• ~ ~: ~ ~·:· -~~, ~: _io -~~ "';~f 75 RESULTS Utili.zation of the food in the habitat With the exception of the last t'ltvO weeks of July, which ·corresponded to the vJarmest t1rvo weeks of the summer, aquatic organisms constituted a higher percentage both numerically and volumetrically of the diet of the fish than aerial insects (Fig. ·23). The aquatic organism~ made up a . ' greater percentage of the diet inlAugust and early September, . V~$ even though the number of aquatic organisms available ~ . -r..f-.~-. at ~highest in early July, at which time the average number of insects per square foot was more than twice as great·,as at any other time during the sunimer (Table I, Appendix 2). Eph~meroptera -Mayfly larvae -.;vere the most numerous organisms in the riffles during July and the first half of ·August. Their relative importance declined as the surn.mer progressed, falling .sharply in the first t~o weeks of Sep- tember~ when they_made up only 12% of the organisms present (Fig. 22) • In the stomachs tfiscc mayflies were found only as larvae, appearing in nearly 30% of the stomachs collected. Their peak of occurrence in the stomachs was in early July, corre- sponding to the peak of abundance in the stream. At this time, in the stomachs, they made up. only 6% of the total· number of organisms present; but, even so, numerically they r JO J1Htt puooes = a: 'J1Btl q_s .. q:J = v · sms1=u-eii'..ro pooJ '"- L wL __ 1m l l 60•- ~50 v <( l! ~401 ..,. •t !::30 ~2ol ... 010 ~ ' l. ·' ' J L I J 0-----o ·Digested matter 0 0 Vegetab.le matter x-----X Aquatic organisms Aerial organisms ;.....0-.. ........ ...... ..... ........ ..... ........ ......, p--------o ''o I I I I .I I I I I I I I 0 x.., ......... ......... .......... A JULY B A 8 AUGUST A S E PT. . J L . j l; o-----0 Aer,al organums 0 0 Aquatic organisms ·.,o, / \\ / ' ~ / -, --', o---- .... A J U L Y 8 ' ---\ _.---:0-- A 8 A~GUST ;A S'EPT. ··-...:~-:.'1-z-g_ o::C.'-'-':__;-..::;~;<:_:-.;:-~,,z,~;.~·~"C'-~D!::'~~n.;~D :··~_,::!!:::.'.:-~.:_,mtr·_-~~--. --~--·~ ~~--!ll"t~~!l!" j --J ()'\ J ., ~-..1 _J ~i " t: ~ .. I>' rr:_, 77 .. 1i1ere second in importance. As the summer progl"er;.;sed, their abundance in the stomachs declined, and b.y Septemberthey . had completely disappeared. Volumetrically t.:he ,· irm·.~e~rnrfl'H':~ erf' the recognizable mayflies in the·stomachs had nearly vanished by.the second week of July (Fig. 25) • . .Plecopt~ -Stone fl:y's larvae appeared in small num- bers in the stream in July, when less ·than orie individual per square ~oot w?,.s _found, but th;Y,· increased slowly, reach- ing a maximum in the last~samples collected in September. At that time, they made up more than 15~ of the total number of organisms present, and had a frequency of 3.4 individuals per square foot. In the stomachsi they were the third most abundant aquatic food item in early July and early August, when they made up nearly 11% of the· total organisms present. They decreased steadily through late August, disappearing ent.irely by September 16th. Volumetrically they were the most significant Tood items found 1n early July, a time when their pres·ence in the stream was at its lowest. In peptember, when they were at their maxj.mum abundance in the. stream, they did not occur in the stoma~hs. Stone flies were the only insect~ found in the stomachs in significant numbers both as larvae and adults. Since the adults found in the stomachs had the wings fully· extended.and the body completely sclerotized, it appears ' 30 ~ \ \ \ \ \ \ Q----~..,.o Tandipedidae o o Epheme!optero :><c----~-.x Plec~pl e ro X--·'~ Simuliidae Trichoptero Kydrocorina Hemiptera \ \ \ . /o .... \ / ........ \ / ........ / .... . 25·-\ /' ................ \ / .......... . \ / f 20 VI ::c v ~15 :e 0 ,_ .... z "' c: Ul !0 ~lO ~ z u.. 0 t~ \ / . . \ / : ~ / I \ ;:f l I \ / i \ / I \ / ! \ / I \ / ; / I ' / I \ / ' / . 'v/ I ! f f I l : I I I . I ,, i I ~, t I .................. , / I ,......_ I )( ........ : I 7'\ ... I I I \ \ : I 1 \ \: I I \ { I I \ ' / \ • ! \ I I \ ·/\: \ / I \ : \ I I \ ' :. \ 1/ '( :\\ /I' \ f \ I I ' ' \ 5 f I /' f ' \ II \ . \ \ ~-v---~ , , I . , '' I ,\_... \ ..... ..· \ .~.x I/-" \ --/7 k ; ...... ·· .... ~\ . =~ .............. _ .. l . ~ . ~-. ____ 1·---------~·-~-----L--I . ~'.:::L ___ _ A s A B A J u l y AUGUST s E P T. r> • .r·J_gur(:; 2'-e· Nu.merical .abundance of aquatic; organisms in the stomachs~ A first ha1f 1 D second half of <:;ach :::-to nth. 7(r ~ -., """' ~ _.... __. -" -" -· --' :j';; l; '" --" ~-- m 10 9 8 -- 7 6 0 05'-0 .... ... \ \ \ •. \ \ •. \ ~ \ \ \ \ \ \ \ \ \ \ \ Ne., atomorph·a ---------. He miptero a o Ephemeroptera 0-------o _ Tendipedidae ~----?< Simvliidae ' ~-----X Plecop~era 0,4 ~ ~ ;::, \ \ •. \ m3 ~. b ~-e 2 \ \· \ /--·-·------·--· \ ; X \ . / /~ . . / / '\ . \ \ // '\ .. \o // ' . \ ... ..... . / \ ,"""'\· ,, . /// . . ' ;.;, ·---... _ ..... ,, ---/6----X \ ··•• ' .;:.--·?7 .....•.... .'--.. ' \ ·· .. :-"'-_......-. I/ _ .• ··· ·· .. :-:;.,.... ', . \ __ ...... ,......~ ......... , . /'/ ......... •':--,_, ...... , '\\ --.... ' ,'/ .. ___ _.,___ . ~· ~-~'u-~ ~ . . ---·<-------.,__-==--~~""~ I ~. _:J A J U L Y A S E P T. 3 A · · 8 AUGUS"f F'igure 25. Volumetric abundance of a.quat.ic organisms l·n t'r1P c::t-on,~-c~lco. A--= .p.;-,-.. c·t· 'rlal·" n . _, ,,..., v ._cl ,J.J.-.:>'"' . ..... J-.1.. v J _._¥l_, ;.J second half of <:~ach month. ,_,C.J I/ __, -. --' _.! ~ so . ·probable that the insects v.rhen taken by the fish were hovering over the water. Thus they were. included a:mong the aerial insects. Trichoptera -~-,addisfly larvae were absent during the summer· in the stream·· s riffles. . However, large numbers of in'dividuals could be seen moving in the pools. They began ·to be found in the fish stomachs in early August, ·reaching a maximum in September w~en numerically they were as important as the Tendipedidae, which 'it'lere the most abundant organisms (Figs. 21 and 22). Volumetrically they became .significant during the latterpart of.August, and then declined slowly through September ·as did.most of the other organisms. Diptera-The only aquatic dipterans-collected in measurable amounts were members of the families Tendipedidae, Simuliidae, and Tipulidae. Tendipedidae-Midge flies were the-second most abundant organisms found in the stream in July and earlY August, but became the most numerous during late August and-September. Their low point in relative abundance occurred in late July, then began climbing steadily (Fig. 25), reaching their maximum abundance at the last sampling date. In the stomachs they were the most numerous single organisms found. Their ups and downs follm1ed closely the variation of the population of the s-tream, the one exception :~ -, __ j ~ _;, -~ -" --" L> b l- 1· ~ t r r f ~· t J I l I ~ ~· .. ·. ~·· I ~ ~ ~ I l ~ • 81 being September. At this time·, while. the stream P.Opulation increased, the abundance in the stomachs declined. Tt. should be noted that, although the relative abundance in the stream changed, the absolute numbers of organisms remained rather · constant J Table I, Appendix II). ( - Due to the midges' small size, their volumetric values were not as high as one would expect from the numbers present. l The high volumetric values for late July and late August are not necessarily correlated to the numbers present, but rather to the size of the individuals; during sorting of the late July and late August samples it was found.that the individuals were considerably larger. Tendipedidae were very common in the drift samples collected betv1een 2000 and 0200 from August 1 to the 15th, at times making up as much as 93%·of the organisms collected. In each drift sample these organisms were found encased in a jelly-like substance and entvlined in filamentous algae. . No organisms vvere ·found in the drift samples collected. on September 2 and 16 (Table III). Simuliidae -The black flies appeared in large numbers in the stream in mid-July, at which time they were· found attached by the thousands to the rocks of the riffles, appearing as a black carpet. After emergence of adults in August, the numbers in the stream decreased rnar}{edly. In the stomachs these organisms' abundance followed ~" " """ '"" ..., _j) -~, ""' ""' ...J _A r t ~ h ~' f, t ~ t ~ ~ iii' ~t $2 closely the changes. ·in population. When the Simulii-ds reached their peak i~~~ the stream in mid-July, they also became the most important aquatic food items volumetrically. Tipulidae·-The crane-flies were not·as common as the other previously mentioned organisms either in the stream or . ( in the stomachs, occurring in on;Ly 2.9fo of the latter (Fig. 26). They were found singly, contributing little to the bulk or to the ·percentage of total numbers in the stomach~ consequently they are interpreted as being;~ncidental food items. Their scarcity in the samples may be due to their habits of living underneath rocks a.nd in crevices. Hemiptera The order Hemiptera was represented only by members of the. family Corixidae. These water boatmen 'II'Tere never found .or seen in the stream, but did occur in 14.7% of the stomachs, and contributed considerably to the bulk in late August ·a·nd September, a time when the absolute number of organisms in the stream was decreasing. Volumef'trically, at this late time, the Hemiptera became the most important food item. Acarina -Water mites (Hydracarina) were never collected in the stream, however they were numerous in nearly halfof the stomachs~-The species collected were anatomically adapted for swirn.rning, having long oar-like appendages. At least four fre~ living species were represented, but.none showed signs of being digested. 1 "-, " J --:' ~ -' """" ~ Figure 26. Frequency of occurrence of aquatic organisms in the stomachs. ;:... "• !fM ![I "'· "ll :r • 3 "V • ... 0 ~ :::t • 3 ... :r 0 ~ ~ :I c.. "ll Ill c.. "' 3' ~ 0:. Q • :::t -a a ... Q ~- . I r----:-·-. ---· . 0 I F REQUEMCY OF OCCURRENCE I I l I -'. I r ~ I I r J - l I_ -il.' l L 1 > ~~ --, ~ _. .......il .... __. 1 ~·· ~ l:~ s f\ i. l r [ w 1: f ~ t ~·· ~ ~' ~· t ~;, ,, l ~y ;>;: :~, ~ !f ' f~: 84 Annel_lida -Segmented worms were moderately abundant in the stjz~am, 'reaching a maximum abundance of 10~ of total organisms in the stream in early August, and then diminishing slmvly. They were found in interstices between and under rocks., which might . explain their absence from the fish stomachs. They occurred in only 4.4% of the stomachs, and always singly. · Nema_tomori?.,ha -27. 9fo of the · ~tomachs contained large numbers of long, thin, brownish worms, -a sample of which \'ITere identified as Nematomorpha of the genus Gordius ·or Paragordius. The highest incidence of occurrence took place in early · July, when they were fo~nd in 87.5% of the stomachs. By late July the incidence dropped to 75.0%, by mid-August down to l2.5fo, and by September down to 9.0%. During July not only was the frequency of occurrence high, but the number in which the worms occurred and the bulk v1hich they occupied made them the most important single item in the stomach. Aerial organisms As previously mentioned, the terrestrial insects ~ade up an important part of the grayling's diet in.the early part of the summer. During rainy periods, flying insects were conspicuously absent from the stomachs, and by September were nearly non-· existent. r ·sqoBmo+s aq+ UT sursTueJl..ro TBT .. tee JO eoue..r..rnooo JO .Aouenbe..r,B: • L'G e..rnJlT.!I r u u I -- '~~~ ·." J ··f I .-- r I I r - I-- I '- ·~ '-... -, %0l » -i m 0 c: ·;, m ~ z ·; r n 1 -< ~ - '" 0 ··-~ ... -~ 'l{, Ol 0 ~ n c:< n '1 c: .! ~ 1 "' -~ m ...___ z ~ n ' m ~ '--:~ ' -'lr.O£ ?~ ~ ,..-- ··l ' ., ~ . ~ ··~ .~ -" 0~ . '-- ....__ - --%09 -~ ----, --, --" _,; d _. 86 The main orders represented in the stomachs were the Hemiptera., Cole'opte-ra, Plecoptera, · Diptera, and Hymenoptera. Hemiptera occurred in 17.6% of the stomachs (Fig. 27), and were represented primarily by leaf hoppers, family Cicadell.idae, which presumably dropped in the water from ( the banks. These organisms 1qere also present in some of the drift samples. Coleopterans were found in 26.5% of the stomachs. These occurrences were primarily from fish collected in July, vri th a few individual remains scattered through August. Diptera and Hymenoptera were the most frequently occurring terrestrial orders, the former being found in 58.8% and the. latter in 48.5% of the stomachs. Diptera were.represerited mainly by the families Culicidae, Tendipedidae, Tipulidae,· and Simuliidae. No attempts were made at quantitatively breaking down the aerial dipterans into families. The Hymenoptera were represented primarily by the · Vespidae, Ichneumonidae, Apiidae, and Formicidae. Members of at least three species of ants were signifi.cant, espe-:- cially in July and early August when they were swarming. -, -, " -' "' b. --" ~~· L----E r i' ~;- ~;·. 87 DISCUSSION The stream grayling is primarily a surface and.mi~-depth feeder. Thus, an understanding of benthic drift is pertinent in discussing :their :food habits. Not until recently has the importance of drifting. benthic organisms been realized in measuring the productivity of streams. Muller (1954) found that drifting organisms made " up a significq.ht portion of the diet of trout. Waters (1961) noted that the total daily drift over an area was many times larger than the standing crop on that area. This could indibate that a fish can be somewhat independent of th~ productivity of the area occupied, and couid rely ~nstead on upstream production. The occurrences of benthic drifts in streams hav.e been described by Muller (1963) as being primarily due to behav- ioral changes in activity on the part of the organisms involved. Typically.the drifts occur at low magnitude during ~he day, vdth high peaks at night (Muller, 1963, \va~ers, 1962). Hm'lever, .recent studies have shown that a considerable variation in times of drift can occur even between related species located in different areas. Waters (1962) found Simulium drifting at a constant rate through the day in the midwest, Pearson and Franklin (1968) found that Simulium exhibited two daily high pea.ks of drifting, at 1600· and· 0200, in the western United States. ·The drift samples collected on McManus Creek -, --"" '--' t'-·- t ~' ~~ c r-·' (~; ~,, J' (, f~ ~: f: $8 were too few to allow the determination o,:fdrif~ing,£imes, but they did show that total drift decreased markedly as . fall approached, nearing zero by September (Table III). The relative importance of the,aquatic organisms found in the grayling stomachs (Fig. 28) follow closely the abun-• . . dance of organisms in the drift samples, more so than the abundance on the stream bottom. It is interesting to note ' the large difference bet\veen the abun<;lance of organisms found on the stream·bottom, and in the drift samples (Table III, Fig~ 22). For example, mayfly larvae and midges were very close in stream bottom abundance during August •. However, in the drift s~mples midges were more than 10 times as abundant as mayflies, a 'ratio·which is also displayed in· the relative importance of organi~ms (Fig. 28). In attempting to find a way of measuring the relative . importance of the food items, no single satisfactory method could be found, since each and every one appeared to be biased toward certain organisms. By plotting the frequency of occurrence, numerical abundance, and·volumetric abundance in the manner previously described (Fig. 29), I felt that the· bias introduced by one method would offset the bias introduced by the· other. Thus the bias introduced by the volumetric measurements favoring the larger organisms would he offset by the axis on numerical importance which favors smaller organ- isms. The results obtained should give a more iealistic S9 --., TABLE III. RESULTS FROM ANALYSIS OF DRIFrr .SAMPLES ----------------------------------~--·--~--~--~~~~~--~---------- DATE _.TIME ANN. EPH. PLEC. HEMIP. TRICH. TEND.-HYDRO. TIPU. C.GOGNA'r. ( 1968) . -~. - 8-3 1200 24,.00 1 5 4 --220 4 8-6 1800 --3 3 --1 65' 1 2 2200 13 4 ----133 2 8-14 1500 ---------'-1 -, 1800 --.1 ------14 1 2300 ----2 -----13 8...,.21 1400 ----1 ----21 9-2 0800 -- -- ------2 ~ 1200 ----1 ----3 1600 ----. -- ------ d . 9-16 1000 1200 1600 1800 ~ --, ... ---' _J r f: .. ' F -" ~. " ~ t~ I ~ fi; '-I ,_ ' I f ,, r 1 lil'' ' -' ·~ ~- r.-: -:::' I -, ·r.t. d :f:;. :~' _. << ~' -' w v z ,;{ ... 1:£ 0 ll. :i ·-... 0 X ..... a z 17 12~ I I l! J I ld·- 9·- -~L-~.i .. ·----=-;.::..~, ~ T,;nd. Sirnuliid. Pl.ecopt. T.-ichopl.ifydroca;. F.phem. H.:mipl. ·17; gurA 0 8-·-__._. -::> .. ..; ~=... ~ Relative importa.nce o.f .food organisms in the grayling stomachs according to the triangul::yr· o.::'ea assessment method. See tex~ and Fig. 29. 90 " ;co -, ... -~ d __. -' ---' ;: ~- ~· t ~-~ i, ~ .::.....:; 3 FREQUENCY OF OCCURR~NCE 60 --------Tendipedidae -------Hydracarina ,j \ I \ I \ I \ D OE h. . \ · · P emeroptera I I Simuliidae *----..,..-)( Plecopte ra . e . o. o o Hemiptera ---,-•---------Trichopt•ra fO I \· I I \ I \ \ \ \ I i \ I I . \ ' \ \ ,. \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ ' \ \ \ \ \ \ \ ~ \ ' \\\ \ \ \I . \ . \ \ 1.-\1 \ \ \ \ f·. \ 1 I \ \\ \ . X -ri.L:LIL.. I I lob.k.J'. . ....0...,..~':--...:...;_ __ .....L, ' 2 10 20 'r. OF TOT A L 8 U L K . ~ OF TOTAL NUMBERS Figure 29. Triangles obtained by plotting frt3quency, numerical values, and volumetric values. .. 91 -.. _.,; ' ~ -.. ~, _ _; -~ -->' --"' -" -' t I. t l ! I· !: ~ ( ! u: t ~· ! I }; ~. ~·· f f ! i<'· 11··.: -·: c· 92 picture of the role of the various food items. The results obtained by this method agree closely.with the results obtained from the drift samples, and the latter I feel are good indices of the food available to the grayling. .The Tendipedidae and Simuliidae.were the most important aquatic food items utilized. Tendipedidae were also the most common items in the drift samples, and for the last half of the summer were the most 1aumerous benthic organisms in the stream. Their frequency of occurrence in the drift can best ·be accounted for by their habits. Tendipedids occurred primarily in close association with algae and organic debris and when these drifted the midges were carried along. Anderson and Lel;.mkuhl (1968) found a similar situation in a woodland stream. Because of the midges' habits of living in the stream's organic litter rather than attached to rocks like the simuliids, five times as many midges than simuliids were present in the drift samples. The conspicuous absence of simuliids in the drift samples from McManus.Creek is perplexing, especially so since they were occurring both in the stomachs and in the Surber samples in limited amounts. Perhaps.the simuliids displayed periodic .drifts which did not correspond to any of the tiines sampled. ·It should also be noted that the first drift samples -v,rere collected at the time of the simuliid population low. · Mayflies and stoneflies were restricted to secondary positions of importance as food items. The_ Epberrieroptera ~ ---.; -., -.. -~ ~-' ., ~ "-' ~ r ~· [ [ [ ..... ,~ u [ [ ~- l_;::, 93 for most of the summer v.rere the most numerous. qrga,pisros in _,v, ~-' •: I, '• • • ·• • • ' ;. • the stream, but oc.curred in low numbers both in the grift samples and in the stomachs. This I do not believe tobe an indication of selectivity by the fish, but rather to-be due to the drifting habits of the organisms. _Some mayflies drift readily, but others do not. Anderson,and Lehmkuhl (.1968) found that of 8 species of Ephemeroptera present in their study area only two contributed· to lthe drift. The others _remained firmly attached to the substrate even during high water periods. -The stone.flies turned out to be more important to the ·latter were .much fishes than the mayflies. _Even though the . I ( . morernumberou'$)-'in the stream, in the drift .... samples they ~ ·' 1n equal numbers. ~uch rlata lead one to speculate that, to the stream inhabiting grayling, the benthic organ- isms present in the drift are the only important ones as aquatic sources of food. The low occurrence of segmented worms in the stoma-chs can be attributed to their-habits. The worms were always found in interstices b~tweeri ~ocks and never in the drift. The Trichoptera larvae were never numerous·in the riffles but were cominon in the pools. As food.items· they -did not become important until September, a time v1hen the occurrence of the summer aquatic food in the stream_declined sharply. The manner 'V'rhich the cadd·isflies were fed upon was -" __, _. _c; -' -' 94 not observed. However, the lack of drifting{9-:UI1a .. and the · observed sluggishness of the fish at this time of the year . allow for some speculation. The fish in the fall spent much of their time close to the bottom. It is conceivable that at that time, due to the unavailability of summer food,.· they ( switched to feeding on the slow moving organisms inhabiting the substrate of pools, in this case the caddisfly larvae. ' I think that the caddisfly ca~es are also responsible for the increase in occurrence of sticks and stones in the stomachs collected in late August ·and September, as illus- trated in Figures 23 and 24. The hairworms (Nematomorpha) present a problem because their role in the stomach is not known. No mention of their occurrei).ce in grayling stomachs was found in the literature. They may have been food :i terns, but they were never seen or collected in the stream. Pennak (1953) and Morgan (1930). describe them as free living, but ·spending part of tbeir .life cycle. as parasites in other organisms, primarily insects. No mention of hairvwrms being parasitic on ·fish v1as found. It appears likely that the nematomorphs entered the fish via some of the insects eaten which had been parasi.tized. Since some of the hairworms showed signs of having been partly digested they were included as food items. rhe water mites (Hydracarina). are known to occupy the quiet pools, where they crawl about the substrate-and brush --, d --, ~ F L.~ ~ b. L, [ l r - l 95 (Pennak, 1953; Morgan, 1930; .and Bak,erand Wharton, 1952). They are not l,\nown to Sl"lim, in rapid current. . A,ppa,rently thos~ mites eaten by the fish had been .swept out of the pools by the current. Although·these organisms had a high frequency of occurrence in the stomachs, volumetrically ( they were insignificant because of their minute size. ·The importance of the aerial insects should not be. overlooked, as they accounted for,at least 40% of the.total organisms eaten and in the early part o.f the ~mmmer accounted for as much as 65% of .the total. Volumetrically their importance was gone by August. By September aerial insects v;ere scarce in the environment. Those seen 1<vere mostly adult forms of aquatic Diptera. From the data collected, one can conclude that duri1~ the summer the-arctic grayling is totally dependent on drifting organisms and flying inse~ts. As fall approaches the fish become more dependent on the benthic drift. Due to the lov; benthic productivity of sub-arctic streams, the fish do no.t show selectivity in choosing their food. This follows Ivlev's (1961) findings that, as the availability of food decreases iri the environment, the sele.ctivi ty of the fish decreases accordingly. --, -, .... -, ~ FEEDING BEHAVIOR In the section on territoriality and social hierarchies feeding movements were discussed in relation to the area occupied by the fish;' the feeding range. In this section the fnovements concerned with the actual taking of .food will be discussed. • METHOD OF. STUDY Observations of feeding behavior were gathered by means of films, sketches, and with the aid of a glass bottom bucket. Movements about the home center .Nere sketched over ten minute periods on papers marked off in ·root intervals • In order to determine if learning took place, various items were dropped in the pools and attempts to ingest the items \"Jere observed. These i terns included crushed mosquitoes, hookless flies attached to 4 lb test nylon line, and spruce . needles. In order to determine the feedfng pe:('iods of the fish two methods were employed. One consisted of testing the catchability of fish at different times of the day. The other consisted of cour1ting the number of rises seen in a pool over a set time. The pool directly below the diversion dam, l'lhich Schallock (1965) estimated as containing over 2000 fish, was used for the catchabili ty study. From June 2·4 until . 96 -o L L [ [ [ [ [ [ I' L [ [ [ r- L 97 September 10, every Sunday, when possible, the pool was fished for five minutes every hour with a brown hackle dry fly to test for surface feeding, and for five minutes l-fi th a No. 1 weighted ·Mepps spinner to test for underwater fe~ding. The number of fish captured and those seen attempting to'take the lures were noted. It became apparent by mid-July that grayling \'Jere capable of recognizing unpalataqle food,items if presented often enough. This was feared to be introducing some bias in the catchability study .. This would be especially true toward the end of the day, resulting from the fish recognizing the lures and avoiding them, even if stil,l feeding. Another index was sought which would not disturb .the fish. The method used was to count the number of rises occurring over a five minute period in a pool having a relatively stable population. For this, pool CC was used. In discussing the feeding behavior I have arbitrarily divided the fish into three main categories; Group A consisted of fish over 23 em in length, Group B consisted of fish between 15 e1nd 23 em in length, Group C consisted of fish under 15 em in length. ; The reason for dividing the fish into these groups was that there appeared to be different behavioral feeding patterns among fish of different. size. It . would not have been feasible to catch and measure each fish in the pools \lfi thout risking the possibility of' changing the fish's future behavior, so the lengths v1ere estimated from above the surface of the \·;ater. ,__., L~ '--" r- w r:-:.:! L. I' L [ [ [ I C..l r·· u i LJ r • i L ~-c L 9S RESULTS Qr~g£ A - These large fish occupied the deepest portions of the pools, staying close to the bottom. Observations with the -g~ass bottom bucket indicated that these fish, even though ( . appearing motionless from the surface of the water, were in reality constantly moving from side to side-as if buffeted by the current. Actually, however, much of this movement wascaused by the fishturning to face drifting particles of debris. Through observations, it '\vas· fou11d that most of· the movement of the fish involved the taking and investigating o.f food items, the only exceptions to this being themove- ments connected with territorial establishment and defense. Three diagrams were made of group A's movements over ten-minute periods. Counts of the number of dashes.made by the grayling showed that 21.9% of all recorded movements were forward movements. Of these, 57.1% were under 30 em from the territorial center, and only 19.0% extended for over 60 em. Of the backward movements, which accounted· for 7S .1% of all noted, 64.2% were less than 30 em in extent, and only 5.7% extended for more than 60 em (Table II). All of these represent feeding movements~ Defensive movements were purposely omitted. ' -., -, _. --' -, ~ .. '-" _c,j _. 99 'fhe movements mentioned were primarily dashes ~ade from the home center to investigate possible food ite~s drifting downstream. Each outward dash was foliowed by a return to the feeding center by a different route. Occasionally seve_ral items were investigated before a return to the home center was made • . These_larger fish occupied positions at the head of the pools, and-seldom investigated fqod items ahead of th~m, but instead turned to face the oncoming object head on. If the object was at mid-depth, a short dash was made as the i.tem passed by, and the item engulfed. For food-items that were approaching close to or float- ing on the surface of the water, the fishes turned-upward so as to keep a constant angle v'li th the food~ In so doing he rose slowly and drifted slightly back1.vards. The fish kept on rising until the food was directly overhead .. With a quick lunge and S1.virling motion the food was taken into one corner of the mouth. As the swirl continued the fish·dove vvhile also moving downstream, making a lateral loop before returning .to his home ·center (Fig. 30) • When large numbers o.f objects vvere .floating .dovm, the fish often. stayed in the mid-depths, and actually moved forv1ard to intercept and · investigate the particles. A characteristic of-the larger fish was the deliberate- ness of their movements in taking food. They rose slowly, 'tc UNIV. OF ALASKA LIBRAR'i1 "U0~1BUB1dxa P811B~ap JOJ ~xa~ aas ·y dno.x~ iq s~uawaAbw ~u1peeA ·o£ aJn~1j ----·--------·----·-------, res..toQ (E) N•a 111. (a) ~-· := --~. "'* ~ "-- (P) (~) '--- ~ -~._, ~· ~ ----·-o • (q) (B)' . ~ -~ ~. ~ ~ ~- oor ' -, ' ..., --' ~ ..... -:::1 ~--..l r -~- 1 ' u ; l --I L..; L l L I r·· L !\ r·. 101 took their food slowly, barely breaking the surfac~·or exposing very little of their back, .and sinking down slightly . . faster than they rose. At no time, over a two summer period, did I see an adult grayling clearing the water while attempting tri take an insect. . . < Crushed mosquitoes, hookless artificial flies, .and spruce needles were dropped in the pools. The grayling re.fused to rise or make investigC}tional movements when the spruce needles were dropped. However, they rose and took the dry flies which they mouthed and dropped several.times_before allowing them to drift backl"lards to where other fish took . them. Usually these larger fish rose to no more than two casts. After a two hour interval the fish rose only to the first cast. When mosquitoes V·.rere dropped the fish ros·e readily to them. However, many drifted past the first fish and were taken by others • GrouE B Observations with the· glass bottom bucket indica-ted that the fish of Group B were in continuous motinn, much more so than Group A. lt,i ve diagrams of the activity of Group .B over· a ten minute period indicate that there was a 58.1% increase in activity over Group A. Also, while the ratio of forward t.o backward dashes among the larger fish was approximately 1:6, those of Group B changed to 1·:2 ratio, indicating a noticeable increase in forward movements (Table II). --, . ., _,'i ..., -, _. ~ ' ~ _j ---, ..... _j --' 102 When dropping artificial flies in pool AA, the following results were experienced: #1 took the first four flies, #2 took 8, v-;hile #3 took the remaining J. It shOuld. be noted here that the first four were taken by #1, and then he refused to take the remaining 11, although on the 7th, 8th, and 9th he did rise close to the fly and then sank back. The 12th, 14th and 15th. were taken by #3. With the spruce· needles, #1 c1ame up and took the first four, #2 took the 5th, 6th, 7th, 8th, 9th, 11th, and 12th; and.#3.took the lOth, 13th, 14th and 15th. In:eachcase the spruce needle was taken under, mouthed, releasedr,·mouthed again, released; and often mouthed several times. After the last mouthing, the needle was allowed to float downstream where it I"Jas picked up by other fish. App~rently spruce needles were not edible. A week later the same experiment vms attempted with the spruce needles but all of the fish refused to touch them. The manner in which the food was takenby<uhis group also varied :from group A. Although.the rise and·<b'e.ckward drift was com.rnon among this_ size group, it was executed much· more quickly. In pools such as CC, where the -#3 position appeared to be often in dispute, the food Has often .taken by quick. forward dashes. vlhen the food was ,taken in this manner, the fish v10uld move upward and forv1ard quickly, take the food in on on:e side of the mouth, -and in so doing expose nearly half -., ~ .., J -' -c:o --' ...... ----" '---"' 103 of the body above the surface. While still moving forward, the ;fish would dive to the bottom and then retu~n to the f'eeding center. Jumping clear of the water, as the food passed overhead, was common in fast current. and 'ltvhere considerable competi.tion ( existed among parr. In these cases the fish would rise rapidly, at approximately a 110° angle to thebottom, grasp the item and clear the water, bending the body in mid-air with the head usually facing downstream, and_ diving quickly for the. bottom (Fig. 31) • This type of _behavior. was especially common on very-bright days • A third type of· feedir1g displayed by this group-·-was one in 't"lhich the parr v'l/ould. clear the water, strike the food item, causing it to sink, and then pick it up on the wayto the bottom. This type ·vv-as found tq occur both ahead of and behind the fish's station in the pool. This behavior was common in the evenings, but occasionally it was -also noted at other times of the day, and no conclusive reason for its occurrence 1.vas determined. While this: type of behavior was in progress in a pool, all of the parr feeding at the time were employing the same: method • Parr feeding at night were found to exhibit_ the deliberate rising exhibited by adults-. They did not emerge from the \lvater, but barely dimpled the surface as they ingested th~ food item. l -j J J 'l J J J J 'l I = L L ".- L . ,........ L -iror ......, 1 -~~ :.d ....., = ~ ~· ! '--' l r" L~ 105 Group C · These smallest fish were ()bserved primarily at pools AA and CC. Although th~y_were distributed primarily at the posteriorend of the pools, individuals were-occasionally found in other portions as well. This group appeared·much ( more active than the other tvm, but no diagrams .for them. were formulated. The b_ack1t1ard ·and forward dashes appeared equal in intensity, as the fish chli.sed everything that·. moved with the current or against it. Territoriality among these fishdid not ·appear to be as marked as among larger fish. This was especially true in pools. where fish of all sizes Vvere present. In these pools the young fish 1r1ere .found only a fevv inches apart and competing for food items which escaped the larger fish. As an i tern drifted dovmstream tvm or three fish would dash fort"iard. toward it, and the first one to reach the ·item .vmuld mouth it. If palatable it was ingested. · If not it was re- chewed several times and .spat out. Eventually it was •. allowed to float backward to v-1here .. the other. fish repe~:ttecl tl1e same feeding movements. These young fish consistently broke the surface v1hile feeding, often leaping clear of. the water. ·They '\orere the only fish.which were regularly seen chasing flying insects vlhich vlere up to six inches above the surface, leaping _and taking them in mid..,..air. In one incident, a four--inch fish ~ ~~ _;; ' _; ~ _, ....,; _.1 _..,;; was seen to.chase a fl.ying cranefly for eight feet, jumping out of the water four times in the process.and ·capturing the insect on the last leap. 106 During the. chas~ng of food i terns. close to the surface, the small grayling often infringed on the territory of larger fish. In this type of relationship, territoriality appeared to break do~om for a few seconds or for the duration of the chase. The intruders were not cha"'~ed out of th~ .territories· and returned by themselves to their respective ar.eas • . Young of' the year, during the time thatthey.spent in the very shallow pools, were not seen feeding. Brown (1938) and So!Th11cimi (1953) mentioned that the fingerlings in hatcheries were feeding on plankton~ and perhaps this is the case in the s·tream also. Feeding intensities In late June the water ,.;as high, turbulent and slightly murky. No fish.could be takenwith the dry fly.· However, the spinner· fished deep and slow d:u.ring,;the day consistently caught fish. For the first two weeks of July, fish-could be taken .at a constant rate at . all. hqurs of tne d9-y Yli th .the weighted spinner. With the dry fly, fishwe:re taken only between 1000 and 1800, reachip,g a maxirp.;wn at 1600 (Table 1 Appendix III) • .. For the last half of July the number of fish taken by· both methods remained constant fromO()OO to 2200. " -, ~ .... _.) -' L.i -" """' ..::._, 107 Through August the fishing success steadily declined. By dry fly, f1sh were taken only betweenl200 and1900. With the spinner, fish .were caught, only>-hetween lOOO·<and '·1800, and at. this time the spinner's effectiveness ·was much·poorer than with dry-flies. 'In September the catchabili ty of the fish dropped to the interval bet1"leen 1100· and 1400 on both the dry .flies and ·the spinner. • By counting the number of rises per five minute period in vool CC, the follmving results were ohtained •. The lowest surface feeding intensity in July 1r.ras noted· through'· earlY:' morning up to mid...;day, increasing slowly ·as evening ;3.pproa:ched, the maximum occurring betv.reen 2200 ·and 0200. Once the ·sun rose over the horizon the fish stopped feeding almost complete- ly. During the maximum intensity of rises, the intensity per five minute period 1.-vas approximately twice the intensity value for mid-day (Table I, Appendix II). In August the maximum number of rises appeared again between 2200 _and 0200. The difference from the previ;ous month ·lay primarily in that the rise intensity bet.weeri noon and midnight increased sevenfold, and the-midnight maximum.was three times the JuJs maximum (Table II, Appendix III}. · Data obtained on Septemhe·r 16 indicated that the feeding periods had changed from the previous months. From 0630 to 1030 no rises o.ccurred. From 1100 to 1500, surface feeding became noticeable but not to the extent seen in the previous -, ' " 9 "'" -., J "" ' -' _ _.;j ._. _Jl -' '--" 108 months •. BY 1500 all activity had ceased except for qccasional rises. After 1800 .it became ·too dark for obse:cv.q.t;,ion. During the early evening .. of September 16, between.1630 and 1800, a mild emergence .of mayflies was in progr~ss over the pool, but this event failed toincrease the inte11sityof . ( r1ses. Ob;servations vtere made on Ju~y 3 and Aug1.1.st 2t which were rainy days. Visibility was ppor due to mq.rkiness, but the fish could still be disce!"ned lying close to the 'bottom. Th¢y did .not appear to be feeding. Attempts,at coup.t;\ng . -· ' . . . ·: :<'···'·, rises failed because if any rises were taking place .• they could not be separated.from dimples causedby~he rain •. Fishing on rainy days proved to.be an arduous task, as flies had to be passed over a fish repeatedly in order to persuade it to strike. ___j --' ' ' _. ~ -., -' _j ""' --' -' -' 109 DISCUSSION The data gathered, sketches, films~ and obseryations, all point to the grayling as.· being primarily'~'a>·mrcl~:de}Yth .· and surface feeder. No evidence of active seq.rching on the bottom was obtained, .and the data from 68 stom.ach.contents · ( support this. The grayling is also a visual feeder. As an item is sighted, by moving from side to s:i!de the fish is able to bring the i tern into the narrov.r cone ahead. of his snout where binocular vision· is· possible (Curtis, l949). Once.· .both eyes a_re fixE!d on .the object, better estimation of distance is pos?ible, and the \"lhole animal moves so that the food item remains within the srnall cone at-a constant angle to the body. The larger .fish, being more deliberate in their move- ment, are more efficient in takingfood items from the surface of the vmter. Since there is no competi,tio:q. :from fish ahead of. him, the larger fish can sight· fin object, and have enough time to intercept it at the food's·shortest distance from the feeding center. The higher interlSi ty of· mov.~men·t shpv.red by the srnaller fish appears to be due to the more intense competition withwhich they are faced. This also explains the difference .in the ratio.of forward to backward dashes shown by Group A ~ -, -~ ~ ~ =c ~ =' _ _. ·--' -" 110 .as comps;tred to Group B. ·The former have no competition tihe~d~or to the sid~ of them, so they can wait until the ·.rood is very ·close before rising in the wat-er column. The smaller fish.on the other hand, inhabit areas where compe- tit~OI:l for each food-item. is steep. The only food items ( r~a,ching .. the. posterior of ·.the pools are those v1hich pass by the larger fish. This· results in the pq,rr,actU:alJ,.y ch~s.ing food i t.ems as soon as they are sight.ed. The· disparity in availab;i.li ty>of food betv.reen the ups~ream and dovmstream portions of .the pool raises some questions. as to the validity of drawing conclusions on learning from the experiments where the different items (hookless flies, mosquitoes, and spruce needles) \'!ere dropped in the pools. It is true that the smaller fish appeared; slowe.r at recognizing inedible food. But is it that tb,ey are slower' in recognizing the i tern,· or is it·· that the competition and scarcity of food force them to snap at anything that is drifting by? I suspect the.latter to be the case. The abandon with which the parr.attack the food· items may also be due to increased compet:i,tion. In·· pool AA, for example, the· largest fish were approximately 185 rmn long, and,thesebehaved like the fish of Group A in pool CG, which contained fishes of all sizes. Thus it would appear that the positions occupied in the pool, rather than the --, ' .., ., d ' --' ~ ... ----, _. -' --' r 111 size of the fish, determines the type of feeding.behavior exhibited. ·.As shown in Tables I and II (Appendix III) thef~eding periods change as the su..11UTier progresses. During the .. end of J·une 1 ~Then the· water ·was ·high·a:nd.;murky, fish could be· -~ . . . taken.during most of the day, but only· close to the bottom. As the summer .progressed the fish could be. taken much nmre l read:ily on the surface· than near the bottom for most of ·the day, with·the peaks in feeding intensities switching from mid"'"'afternoon to midnight. In August the intensity of the midnight feeding activities increased drastically, until the midnight illumination dropped to less than 10 foot candles in mid August. By September feeding at night stopped comp:letely, and· the feeding acti vfty took place only dU.ri1g . a.short·period at mid;....day. At lower latitudes Hoar (1953) found that trout also did not feed at night. From· July 16 to Septemb-er 2, the data from the catch-· ability (Table I Appendix ITT) and the rise.experiments (Table II, Appendix III) do not seem.to agree; ·Since they .were taken from areas 'having different characteristics, -one a pool with much·'standing water,. the other a running pool, this might account for the differences •. J:·tend to accept the rise count data more·. readily because it \'Vas collected from a pool typical of the stream, and because the fish were not disturbed in any manner dur.ing the counts. -, " -. -" =- ~ _..; d -' _.; 112 During September the feeding period became restric.ted to the time between llOO and 1400, but during this period the. fish fed intensively. However, the fish needed consider- able motivation before beginning to feed. The feeding activ- ity increased in inte:h'sity only after one fish reinforced ( or motivated another •. That is, when one ·rish·began to rise over another, the latter did not move until the first fisp had risen several times, after 1 .. vhich the second fish rose slowly. After two or three such rises it began to be more ·active, and other fish in the pool soon follOwed suit-. Similar behavior was observed in other· drainages as late as mid-October. In the cases mentioned above, it appeared as if the food was not-the·stimulus· releasing the feeding activity in the fish, but rather the sight of other £ish feeding. During rainy days the feeding activities appeared to be nil. These observations are supported by the analysis. of eight stomachs collected on rainy days, and comparingthe bulk contained with the stomachs of ·fi'shes· of:'Simia:ar si'zes collected on sunny days (Table IV). In most cases the total volumes of the stomachs collected on-rainy days were 1/3 to 1/4 the value of the sunny days total, and ainong the stomachs coLLected on rainy days 78 to 100% of.the contents were a partly digested mass indicating· that no recent feeding had · taken place. The stoppage of feeding I attribute to the -1 . _, ~ """) --" ....., -• ~ ·-' ~ _, -" I I t l l ~ ! ! ! >. t. 113 TABLE IV. -STOMACH CONTENTS ON RAINY DAYS AS CONTRASTED TO SUNNY DAYS. . --~~,----~~~~~----~~--~~------~-------------~--~------~~~----- RAINY DAYS Date Length of. Coll~ct·ed · Fish 8/5/68 208 240 180 170 170 147 140 Vol~ o! Stomach Contents (in ml) 0.35 0.40 0.23 0.22 0.37 0.10 0.23 SUNNY DAYS Date Length of . Collected · · . -Fish ...... ·~-·~----~.....,. 7/22/68 ' . 218 243 173 147 138 Vol~-,or· -Stomach-- Contents ( rn·.··1n1 l - ~~~+:- 1·.·-'i"·T-d-·.~,·~'0 2~':28 0~60 0.-48 0.45 -------~--~--------~~--------~------·--------~--~~-~~--~--~----,~~~------~~----- :.:. .. J J J ] J J J ] J J ] ] ] ] J J 114 inability of the fish to see the food i terns drifting down due ·to the roiled water. Anderson, -and-Lehmkuhl { 1J;168) "'found ' ·'· . .·. .· ··. ·. that, during periods of high water and rain, the amount6f drift in a stream increased from five to ten times·the sunny day's normal. This indicates that.the grayling are unable to utilize a substantial f()6d sou'rce available 'du::t:-ing periods of.heavy rain. This tends to support·the hypothesis that·the grayling is a visual feeder. The result of the distribution and ori~ntation pf the fish "in the pools was that they v1ere. distributed so.,,,as to best utilize the food which vvas being· transported into the p~ol by the current. The fish's food habits and feeding behavior both indicate that they are primarily mid-depth and surface feeders. What is even more important, the bulk of their diet is made up of benthic drift. This benthic drift dependency is an interesting adaptation, as it· allows ·f~sh inhabiting low. production streams, to utilize f.ood produced in other areas. Thus in Mcl\1anus Creek' most .c>f' ·the St'r .. E'hirn consi~t~d of riffles, which are areas of very high benthic productivity. (Mo:rgan, 1930}. The fish amassed in the pools w~re able< to util.i ze large areas of production upstream from thein (i.e., all of tJ:le areas bet\veen pools). The relatively small sizes of the pools, coupled with the infrequency of pools in the stream, and J J J J 'I u ] J J J J J J J J J d J J ] 115 territoriality vlhich may limit the number of fish in -the stream, aJ_lovi for su,rvival of the pogulatiqn in the .u,u,p:ro- ductive stream. The within-pool distribution allov.rs for the larg~s~ fish to utilize the incoming food supplyat its ful:Lest. ( Sincethey occupy the portion of th,e pool where the water enters in greatest volume, it appears likely that they are able to utilize a considerable number of organisms inthe drift. By staying close to the bottom they conserve er1ergy since the current is considerably> slower there. o:Be~cause of their sizei it is conceivable that th~ir needed-daily ration is larger than that of the smaller ,individuals, and by occupying the upstream position they are able to satisfy this. The result of the distribution of thefish is that the fish needing smaller rations occupy positions where the food made available is lower. What appears to take place in the grayling pools is that feeding territories are set up ina hierarchial fash:i,on by means of dominant subordinate relationships. "The result is that bj means of behavioral displays a distribution is achieved satisfying the nutritive needs of· the :individuals involved, and a maximum utilization of the resource is attained. A secondary benefit of such a distribution is that it helps in maintaining a healthy stock of fish. The fish in better condition, by means of displays are able to gain J l J J J J J J j J J J J J J J J J ] 116 . a territory where their nutritive needs can be satisfied. V/eak individuals, such as the tagged fishes, . are releg?.ted· to positions where their daily. rations are smc:iller than their needs. The result of this is a poorly conditioned animal whi.ch may not be able to survive the long winter, a time wheh the benthic fauna is drastically reduced. ' l l J J J J ] J J J J J J J J j J J J SUMMARY 1. Tagging and recapture of Arctic Grayling in McManus Creek during.the summers of 1967 and 1968 indicated that an upstr;eam migrati·cm takes place through the sutnrnere 2. A diversiondam locat~d below the junctiort.of Faith and McManus Creeks was found not tp be a complete barrier to • fish migrating upstream during the spring, but may act as a barrier during the summer. 3. The grayling were found to distribute themselves in the summer poois according to size, the largest ones occupying the uppermost portions of the pools, and the'smallest ones the periphery. 4. Each fish in a pool haq a feeding range vdth a range center from which all of his activities began. Since the area v-ras defended by aggressive displays the area defended vvas considered a feeding te~ritory. 5. Each· individual's. position in the· pool·was ·.0btained by behavioral displays between sets of individuals until. one ·emerged as the dominant fish and the other as the subordinate. In this manner a pseudo-straight line hierarchy was formed. 6. In the fall.the fish began moving out of the pools and the hierarchies dissolved. 117 J l J J I J J ] J J J J J J J J J J ] J 118 7. Sixty-eight stomachs were analyzed q.nd from this data and visual obseryation it was deter~nrtned t.b..<:!t the grayling is primarily a surface+-and mid-depth feeGl:er. 8. Benthic drift and flying insects :were found to/ be the most important_organisms eaten. In J:uneandearlyJuly ( flying insects were the most important food i terns, .and the rest of _the time-. benthic drift was the important food source. l 9. In order to determine the relative_importance of various food items their frequency of occurrence, numerical abundance, and volumetric abundance were plotted so that a triangular area was enclosed. The ratio of the various areas was used as an index of relative importance. 10. Behavioral observations appeared to indicate that the grayling is a visual feeder. Stomachs collected at times when the 1vater was roiled substantiated this~ 11. Adults and parr·exhibited different feeding behavioral patterns. The former fed in a deliberate manner; the latter instead fed quickly, often jumping clear of the water :v-ihile taking the· food. 12. The maximum feeding intensity peniods of the gray- ling shovv several variations. In August the peak :v-1as at midnight, and in September the fr::eding period was limi·ted from 1100 to ll,.oo. J l J J u J J J J J J J J J J J J J J 13. It .is postulated that the feeding territories established during the suniil;ler ultimately result in. the 119 best utilization of space for fish having thefeeding habits displayed by the gr~yling. J J J J J J J J J J J j J J J LITERATT.JRE CITED AI+ee, vl. C., A. E. Emerson, 0. Park,. and K. F. Schmidt. 1949. Principl~?s of animaL .ecology~ Saunders,· · Philadelphia.and London. · 837 p. . Anderson, N. H., and D. J~. Lehmkuhl. 1968. ·drift of insects in a \'lood..l;:md str~?m• 198-206. ( Ca tas-t,~;r-oph;i c Ec,plogy.49: Baker, E. VI., and-G. W. Wharton. 1959 •. An introdu'ctionto Acarology. The HacMil1an·company, New York. 465 p. Braddock, J. C. 1945. Some aspects of the dominance-·· subordination relationship in the. fish Platy:E.oecih~s maculatus. Physio1. Zoo1., 48£176-195-.·'···.· ·., ·~ ···.··. Brmm, C. J.D. 1938a. Observations·on.the1ife~hi$.tory and breeding habits of the Montana grayling. Copeia, 1938:132-136 • • 1938b. The feeding habits of the Montana grayling ---'[Thvma1lus montanus). Jour. Wdlf. Jv!gt., 2:135~145. 'Brovm, C. J. D., and C. 'Buck, Jr. 1939. When do trout and ·grayling fry begin to take food? Jour. W'dlf. Mgt., 3: 134~140. . Burt, W. H. 1943. Territoriality and home range concept as applied to mammals. ··Jour. Mammal., 24:346-35.2. ,Carl, G. C., VJ. A. Clemens, and C. C. Lindsey. 1959. The freshwater fishes of .British Columbia. · British Columbia Provincial Museum, Dept. Education, Handbook ·No. 5. 192 p. Cartis, B. 1949. and manners. 284 p. The life story of the· fish: 'his·ntorals Dover Publications, Inc., New.York. Degteva, G. K. 1965. Materialy po biologii khariusa ozera Chi.stogo. Tikhookeanskoga Nauch-Issled.Rybn Khoz Okeanogr., 59:224-226. t Original not. seen. From Bips. Abs., 48:52885~) EgoroY, A. G .. 1956. Mechenie Khariusa na r Angare. Voprosy Ikhthiol., 1956(6):121. (Original not seen. From Bios. Abs., 32:259$0.) 120 J l J J J ] ] J J J J J J ] J J J J 121 . Evermann, B. vJ., and E. L. Goldsborough. 1906 .. The fishes /"' of Alaska. Bull. U. S. Bur~ Fish., 12:219-376. Fabricius, Eric, and Karl-Jacob Gustafson.· 1955. · Observa- /"'"' tions on the spawning behavior of th.e grayli11g ~ 1JIY1!:!al:Lt+s thymallus (.L) . Rept. · Inst. Freshwater Res. , Drottningholm, · · 36:75--103. Frankel, G. S. , and D. L. Gunn·. . 1961. The orientation. of animals. Dover Publications, Inc.; Ne•V'f York. 376 p. Gerk'ing; S. D. 1950. Stability of a stream fish popula- tion. Jour. \vdlf. Mgt., 14:194-202. .....----.---· 1953. Evidence for the concepts of home range I' · and terri tory in stream fisl46s. Ecology, 34:347-365. G;t~eenberg, B. 1947. Some relations between territory, social hierarchy and leadership in the green sunfish CLepomis cyanellus). Physiol .. _ Zool., 20: 26?~299. -~Hoar, vl. S. 1953. Control and timing of fish migration. / Y Biol. ·Rev., }8: 437-452. . · · · /' Ivlev, V. S. 1961 .. Experimental ecology of the feeding of · fishes. (Translated from Russian by the Yale Univer- sity PrSss, New Haven. 302 p.) Kafanova, V. V. 1965. K izncheniyu khariusa ozera Ni'zhnee Kulagash-Bazhi (bassein r. Chylysh..rnana). IZV Ataiskogo Otd Geogr Obshchest SSSR, 5:165. (Original not seen. From Bios. Abs., 98:16208.) Lagler, K. F. 19)6. Freshwater fishery biology. Vlm. C. Brmrvn Company, Dubuque, Iowa. 421 p. /Leach., G. C. 1923. Artificial prbpqgatio:f.l of: whitefish, grayling, and lake trout. Bur. Fish. Doc. No. 949 . . 39 p~ · Leonard, J. W. · 1938. Feeding habits· of the· Montaha gray- ling. (~a1lus montanus Milnes') in, Ford 1akl:!, Michigan~---Tians·:--A.m. Fish. Soc., 68:188~195. Miller, R. B. 1946. Great Bear Lake • Notes on the arctic grayling from Copeia, l946{4)~227~236. · - . 1957. Permanence and size of home territory in ----stream-d1rtelling cutthroat trout. J. Fish. Res.· Bd. Can., 15:27-45. Morgan, A. H. 1930. Field book of ponds', and• streams. G. P. Putnam •· s Sons~ Ne'V'l York. 4h8 p. J l ] J J J J J J J J J J ] J J J 122· . Muller, K. 1954. Investigations of qrganic_ dpi;ft" in north Swedish streams. . Rept. Inst. Freshv.rater Res., . Drottningholm., 35:133-148. . 1963. Diurnal rhythm in "organic" drift p.( __,.......,..,.,...... ,~GammC?-:ctls .:ey:lex. Nature, 198: 806-807. Nelson, P. H. American 324-342. 1954. Life history and management of the grayling in J.VIontana. Jour. Wdlf. Mgt., 18: Newinan, M. A. 1956 ~ Social behavior and. interspecific competition in two trout species. Physiol. Zool·., 29: 64..,.81. Noble, G. K. 1939. The experimen.,tal animal from the naturalist's .point of vie\I'J'. Amer. NaturaJi~t, 73: 113~126. Noble, G. K., and B. Curtis. 1939. The social behavior of · the jewel fis.f!, Hemic;,hromis bim<3.culatus. · Bull. :A.mer. Mus. Nat. Hist., 76: ll-7o. . ·. · · ··. · ··· Pearson, W. D., and. D. R. Franklin.· 1968 •. ~Some fact.ors affecting drift rates of Baetis and Simuliidae in·· a large river. Ecology, 49?76-lfi. Pennak, R. W. 1953. Fresh water invertebrates of the United Ste1te·s. The Ro!jq.Hi Pres$ Compc;t.:rlf, New Yoik~ ?69 P··. . . / J Peterson, H. H. 1968. The grayling Thyrrg:tlJus .!dLvrna.ll;._l,ls . (L), of the Sundsvall;Bay Area. Rep. Inst. F'resh,\'mter Res., Drottningholm1 48:36-56. Reed, R. J. 1964. Life history and migration patterns of 1 f· arctic grayling, T~JI!!al~ ~,..£!-~~ ( f?alla~.), in 'the Tanana.river drain.age·of ·Alaska. ·Research·ReportNo. 2. Alaska Dept. of Fish and·:~·Game. Schallock, E. \'l. · 1965. Inve_9tigations deali11g with / probable grayling movemont j_h rel.atipnship to ;·\\J:ater d.evelopment projects. Unpublished M.S. thesis, University of Alaska. 117 p. Shetter, D. S. 1937. Migrat.ion, grov;rth rate,. and popula- tion density of brook trout in the north branch·· of. the Au Sable, Michigan. Trans. Amer. Fish. Soc., 66: 203~210 .. // Sornmani, E. 1953. Esperitnenti di ·allevamento art:lficiale ~: del temolo ( Th. tr:J.:!!!al~us L.). Boll. Pe9ca Piscic. Idro- binl., 47-57. · · J l J J .-1 u J J J J J J J J ] J J J J J 123 . . Starmach, K. 1956. Rybacka i biologiczna. charakterystyka rzek. Polskve Arch • Hydrobiol., 3:307-332. (Original not seen. From Bios. Abs., 33:8897.) _Svetovidov, A. 1936. Grayling, genus _'!'rry:maJ.J;.us Guvier, /_, of Europe and Asia. Travaux de·l'Institut Zoologique de l'Academie des Sciences de l'URSS, 3:183-301. / Tryon, C. A.· 194 7. The Montana grayling. Prog. Fish. Cult., 9:136-143 • /Vincent, .. R. E. 1962. Biogeographical .. and ecological ,/-factors contributing to the decline of Arctic gray- ling, ThYn:!allus arcticus Pallas, in Michigan and Montana. D:i.ssert. Abst., 23:3059-3060. ' _/ Ward, J. C. 1951. The breedir1g biology of the. Arctic / grayling in the southern A thabaskari Draina,.ge. ·. Unpubl. M. S. thesis, University of Alberta. ·waters, T. F. 1961. Standing crop and drift of stream bottom organisms. Ecology, 42:532.-537. _. 1962. Diurnal periodicity in the drift· of stream ---.invertebrates. Ecology, 43:316-320. \f.lindell, J. T-' 1968. Food analysis and rate 'Of digestion, p. 197-203. In \v. E. Bick?r, Methods .for assessment ·of fish production in fresh waters. Blackwell Scientific Publications, Oxford and Edinburgh. )~'lojcik, F. J. 1955. Life history .and management of the // grayling in interior Alaska. Unpublished M.S. thesis, University of Alaska~ 54 p. ~ L._j ~ c..LJ L.:...J L.J L..J L.J L.J ~'"" r; ~ L!.J L..2J L__l ~ C......J L..J LJ L:...J l,__j , APPENDIX I TABLE I. SUMMARY OF PHYSICAL CHARACTERISTICS OF THE POOLS OF MC MAN{)'$ CREEK. POOL SECTION SURFACE IvrAX. DEPTll TYPES OF BANKS SUR-BOTTOM COMPOSITION OF THE .POOLS OF STRM. AREA OF POOLS ROUNDING ,THE POOLS (Loca-OF POOL (in f'eet) Brush Brush Both Banks tion of (meas. 2 pres. pres. barren % Rubble % Gravel % Sand Pool) in ft. ) (both (one banks)bank) 1 Upper 240 1.5 --X --100 a a 2 II 35 2.0 ----X 100 a a "3 II 400 1.5 ----X 50 50 0 f'-1 !\) +-4 " 150 2.5 ----X 100 0 0 5 II 200 1~5 --X --.. 100 a 0 6 II 140 2.0 --X --100 0 a 7 II 24 2.0 ------100 a a 8 II 180 2.0 X ----100 a a 9 II . 198 4.0 ------100 a a· 10 " 2.70 3.0 ----X 10 80 10 11 II 280 3.0 --X --. 75 5 20 12 II 75 "1.0 X ----100 a 0 13' II 180 5.0 --X --10 15 75 14 II 300 J.O ----X 75 20 5 L..JL..JL..JLLJL.JLJL..JL_]L.Jc:.:.....:Jc.JL.J.JL..JL.Jc...JL..JL..JL...:_]L...J APPENDIX I TABLE I. .Continued. 15 " 450 3.0 --X --50 35 15 16 " 216 2.5 X ----50. 50 0 17 " 77 3.0 --X --100 0 0 18 " 120 1.5 --X --100 0 0 19 Middle 275 2.0 --X --70 25 5 20 II 24. 1.0 ---..,. X 50 50 0 21 " 280 3.0 --X --50 50 Q 22 II 150 3.0 X ----85 15 0 23 II 480 4.0 ----X 100 0 0 ... 24 II 350 2.5 --X --60 30 10 25 " 275 3.0 --X --80 10 10 26 " 350 J.O --X --60 40 0 27 II 270 3·5 --X --85 15 0 28 II 220 1.5 X ----100 0 0 29 " 525 4.0 --X --95 0 5 30 II 860. 2.0 --X --95 5 0 1...,.1 . N. VI L,_jc.__;L_]C_jc...::..JL..JL..JL.JL._Jc.:.....JJL:.JL.JL..JL:....::Jc...:JL..JL...J~L..J TABLE I. n " 32 33 34 II II II 35 It 36 It '37 38 39 40 41 42 43 } Lower II " . II .II II II APPENDIX I Continued. 2 97 3 .o --X --3 0 50 2 6 . 3 00 4. 0 --X ---7 5 . 15 10 450 4500 2250 105 220 200 525 . 600. 1500 1000 1000. 5.0 5.0 4.0 4.0 2.0 3.5 5.0 5.0 1.5 5.0 .4. 5 X X X X X X X X X X X 50 25 25 70 20 10 ·. 60 30 10 75 20 5 100 0 0 50 25 25 10 0 90 10 35 55 33 33 33 75 15 10 50 40 10 f-l 1\) ()\ . L....: r : "-----' LJ CJ r· 1 '-----' [_J L...J [_J LJ c.:..JJ L..J L.JJ L_j L...J L_J L_J [_j :....:...J L._j APPENDIX I TABLE II. SUMMARY OF PHYSICAL DESCRIPTIONS OF POOLS ALONG MC MANUS CREEK. SECTION POOL POOL LENGTH DEPTH(arith-SURFACE NO. (arithmetic metic mean AREA mean in ft.) in feet) Upper 18 25.1 2,~44 194.4 Middle 18 38.8 2.•97 661.2 Lower 6 . 51.0 3.79 720.7 No.Vege- tation 31.2% 11.1% 28.6% .. BANK TYPES Vegetation Vegetation on 1 bank· . on both 50.0% 18.7% 72.2% 16.6% 71.4% -G- I-' l\) ...J· [_j L.... L.J u...J L:.J r 1 r , , .....____: L...J L..J L....J c....D L.:..J [__JJ [_J L_J L....J L.J [_j l......:_j L._j APPENDIX I . . T;\BLE III. SUMMARY OF' BOTTOM COMPOSITION OF POOLS ALONG. MC MANUS CREEK. .. ~ RUBBLE GRAVEL % of % of Freq. j· % of · %of Freq. %of. ' I Bottom Bottom· of 1 /, Bottom Bottom· of Bottom of of occur-; of of occur-of· pools seqtione renee J pools · .. ; ·section renee pools I covered covered in ~~ covered covered in cove :red I by' by sec-. by . by sec-. by rubble rubble tion f;. gravel grave~ tion sand I ,'·• • ... ," ·/ I 'UPPER SECTION 77.6% 77.6% 100% . 36 .• 4% 15.0% 41.2% 25.0% ~· 't ' ,. ~~ 26 .6~ j 22 • 2~ .I s:3 • 3~ MIDDLE SECTION 71.6% 71.6% 100% '11.1% I ' I . . . ~. . •. . I··· •"' . ' .. [ LowER SECTION 46.8% 46.8% 1100% 29.0% 20.8% 71.4% 37.1% ~----··------------------~-~-· .. SAND % of Freq. Bottom · of of occur- sections renee .. covered · in by sec- sand tion ( 1_. ·' 7.3% . I . ,J·~9-4% I 5.5% J 50.0% ·~ i-•.1 . ' ·~ . ·l~v ... :. · 31.8% ll5.8% :.~ J ' H l\)' CQ. L__; LJ L_j Ci....J L_j [_J L.J r .,~ L.....-.J L.J c..JJ LLJ L...J L.J L..J C,_J L..J L.J ~ . APPENDIX I TABLE IV. SUIVIMARY. OF THE .WATER VELOCITIES PRESENT IN THE POOLS ALONG McMANUS CREEK • SECTION WHERE POOLS . OCCURRED UPPER SEC. Frequency of Occurrence of Water Velocities FAST MED. SLOW (18 pools) 61.1% 66.6% 38.8% MIDDLE SEC. . .(18 pools) 61.1% 8.3.3% 44.4% LOWER SEC. (6 pools) 16.6% 50.0% 83.3% .. Frequency of Occurrence of_ Combinations of Water Velocities FAST ONLY 16.6% 11.1% -Q- FAST MED.& SLOW MED. & FAST MED. & SLOW 11.1% 27.7% 11.1% 16 .6% 33 .3% ... 22 .2% 16.6% _.g_ 33.3% MEDIUM ONLY 16.6% 11.1% 16.6% SLOW ONLY 16.6% 5-5% 33.3% ~ f-'-1.. !\) '-() \' J APPEfHHX II J TAHLEi I. RESULTS OY ANALYSIS: OF-t;SURBER SAMPLES '•· :·. ··,:~·-~"··· ·.·.;.'4 .. ---~' -.~.---· .. -~~~~~~ . .; .·. ·1JATE . TOTAL J . ,E-t • cot.~ H . ~ p... • 81'. O.RGAN-H ;?!; 0 H . . • LEC!fED ;r£1 r£1 0 0 !!':> d ~ :::::> ISMS: (1968) ~--·::r:: H, r£1 ':;;t ~ :z; p... J :z; p... g::· H H ::r:: ~-H • < r£1 6-1 p... Cf) ~ E-t Ol ... , .. .,.-:,· . 7/7 --4 1 5 J .,. 10 75 3 11 3 51 153 " 3 89 l 1 11 2 52 159 J " 3 53 6 1 45 108 tt· .;.1i,._ 36 14 1 51 "If J 7/8' 3 153 -""!"". 6 89 242 493 ·II 4 9r7 ---55 1 51 3 211 7/20 1 30 3 26 10 1 71 J 7/22 32 1 3 9 1 l~6 " 14 34 3 7 --58 J " 2 43 7 4 56 " 8 17 10 34 21 2 147 J .8/5 4 29 1 7 6 29 2 75 ' tt 2 7 1 1 3 1 15 J " 4 11 1 2 1 9 1 29 II 7 21 2 2 43 l ---77 J 8/12 12 7 1 12 1 1 34 " 6 l 3 10 tt· --16 2 4 12 34 J " 12 5 1 19 1 41 8/18 3 18 7 2 24 54 J " 2 11 4 1 5 23 9/2 1 3 9 4 17 ' J 9/14 8 10 12 5 1 30" 2 68 d J J 130 ] ~~~~~~~~~~~~~~~~~~~\ \ APPENDIX III. TABLE I .. CATCHABILITY OF FISH. AT DIFFERENT TIMES OF THE· DAY DRY WEIGHTED DRY WEIGHTED DRY WEIGHn;D DATE TIME FLIES SPINNERS DATE TIME FLIES SPINNERS DATE· TIME FLIES SPIN1"ERS 6/24/68 1000 l 1 7/21/68 1000 1 2 8/24/68 0800 0 0 ).100 0 1 1200 2 1 1000 0 1 1200 0 2 1400 3 2 . 1200 1 0 1500 0 2 1500 3 1 1600 1 1 1600 0 2 1600 4 '2 1800 1 0 1800 0 1 1700 0 2 2200 0 0 ~~100 0 0 1900 1 ·1 9/2/68 0800 0 0 7/7/68 0800 0 0 7/28/68 0600 0 1 1000 0· 0 f-1000 0 0 0700 2 1 1200 2 1 I..,V 1200 1 2 0900 3 3 1400 1 2 i-~ 1400 2 1 1000 3 2 1600 0 0 1500 0 0 1700 3 2 2400 0 0 1600 0 1 1800 3 2 "' 1700 0 2 '2200 2 1 "9/14/68 1000 0 0 1800 0 0 2300 2 0 1100 2 1 1900 0 0 2400 0 0 1300 1 2 2200 0 1 7/29/68 0200 0 0 1400 0 3 2300 1 1 1500 0 0 2400 0 0 8/10/68 0800 0 0 7/$/~8 0100. 0 0 1000 0 1 1200 2 1 7/15/68 1500 0 1 1400 3 2 1600 1 0 1500 0 2 1700 2 3 1600 l 1 1800 11 3 1700 2 1 1800 1 1 1900 l 0 2000 0 0 -----_______ ._ ________ J 132 APPENDIX III n TABLE II. RISE COUNTS. AT POOL cc. J AVERAGE NO. OF RISES/ DATE .TIIV1E 5 MINUTE PERIOD n 7/16/68 1200 2.4 1400 2.2 1500 2.0 0 1600 3.4 (. 2200 6.6 2300 9.0 J 2400" 7.8 7/17/68 0100 6.5 0200 -G- --· ~-- J 7/28/68 0900 2.4 1200 3.0 1400 3.2 0 1500 3 .l;. 1900 4. 5 2400 8.5 7/29/68 0100 6.0 J 0200 2.2 ________ , __ 8/3/68 . 0900 J.4 J 1200 6.0 1500 11.8 2000 20.0 J 2200 21.1 --"'<!'co-___ 8/21/68 0900 2.4 J 1100 4. 5 1300 6.0 1600 .12.0 2200 27.4 J 2400 28.2 0200. 6.5 J 9/2/68 0600 ~G- 1000 -Q- 1200 2.4 1400 4.0 J 1500 1.8 1700 0.2 1800 0.2 g 1900 ·-G-· ' -'~~--.:.-..;----·--=----· ---....=---====· J J J